MongoDB Documentation

Release 2.4.9

MongoDB Documentation Project

March 18, 2014

Contents

Install MongoDB
1.1 Installation Guides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2 First Steps with MongoDB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3
3
19

MongoDB CRUD Operations

2.1 MongoDB CRUD Introduction
2.2 MongoDB CRUD Concepts . .
2.3 MongoDB CRUD Tutorials . .
2.4 MongoDB CRUD Reference . .

27
27
29
58
82

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

Data Models
3.1 Data Modeling Introduction . . . .
3.2 Data Modeling Concepts . . . . . .
3.3 Data Model Examples and Patterns
3.4 Data Model Reference . . . . . . .

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

97
. 97
. 99
. 106
. 121

Administration
135
4.1 Administration Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
4.2 Administration Tutorials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
4.3 Administration Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223

Security
5.1 Security Introduction
5.2 Security Concepts .
5.3 Security Tutorials . .
5.4 Security Reference .

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

237
237
239
245
267

Aggregation
6.1 Aggregation Introduction
6.2 Aggregation Concepts . .
6.3 Aggregation Examples . .
6.4 Aggregation Reference . .

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

277
277
281
292
308

Indexes
7.1 Index Introduction
7.2 Index Concepts . .
7.3 Indexing Tutorials
7.4 Indexing Reference

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

313
313
318
339
375

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

377
377
381
419
466

Sharding
9.1 Sharding Introduction . .
9.2 Sharding Concepts . . . .
9.3 Sharded Cluster Tutorials
9.4 Sharding Reference . . .

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

479
479
484
506
546

10 Frequently Asked Questions

10.1 FAQ: MongoDB Fundamentals . . . . . . . . .
10.2 FAQ: MongoDB for Application Developers . .
10.3 FAQ: The mongo Shell . . . . . . . . . . . . .
10.4 FAQ: Concurrency . . . . . . . . . . . . . . . .
10.5 FAQ: Sharding with MongoDB . . . . . . . . .
10.6 FAQ: Replica Sets and Replication in MongoDB
10.7 FAQ: MongoDB Storage . . . . . . . . . . . . .
10.8 FAQ: Indexes . . . . . . . . . . . . . . . . . . .
10.9 FAQ: MongoDB Diagnostics . . . . . . . . . . .

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

555
555
558
567
569
572
578
581
585
587

11 Release Notes
11.1 Current Stable Release . . . . .
11.2 Previous Stable Releases . . . .
11.3 Other MongoDB Release Notes
11.4 MongoDB Version Numbers . .

Replication
8.1 Replication Introduction
8.2 Replication Concepts . .
8.3 Replica Set Tutorials . .
8.4 Replication Reference .

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

591
591
609
634
634

12 About MongoDB Documentation

12.1 License . . . . . . . . . . . . . . . . . . .
12.2 Editions . . . . . . . . . . . . . . . . . . .
12.3 Version and Revisions . . . . . . . . . . .
12.4 Report an Issue or Make a Change Request
12.5 Contribute to the Documentation . . . . .

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

637
637
637
638
638
638

Index

ii

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

655

MongoDB Documentation, Release 2.4.9

See About MongoDB Documentation (page 637) for more information about the MongoDB Documentation project,
this Manual and additional editions of this text.
Note: This version of the PDF does not include the reference section, see MongoDB Reference Manual1 for a PDF
edition of all MongoDB Reference Material.

1 http://docs.mongodb.org/v2.4/MongoDB-reference-manual.pdf

Contents

MongoDB Documentation, Release 2.4.9

Contents

CHAPTER 1

Install MongoDB

MongoDB runs on most platforms and supports both 32-bit and 64-bit architectures.

1.1 Installation Guides

See
Release Notes (page 591) for information about specific releases of MongoDB.

1.1.1 Linux
Install on Linux (page 3)
Install on Linux
These documents provide instructions to install MongoDB for various Linux systems.
Recommended

For easy installation, MongoDB provides packages for popular Linux distributions. The following guides detail the
installation process for these systems:
Install on Red Hat Enterprise Linux (page 4) Install MongoDB on Red Hat Enterprise, CentOS, Fedora and related
Linux systems using .rpm packages.
Install on Ubuntu (page 6) Install MongoDB on Ubuntu Linux systems using .deb packages.
Install on Debian (page 7) Install MongoDB on Debian systems using .deb packages.
For systems without supported packages, refer to the Manual Installation tutorial.
Manual Installation

Although packages are the preferred installation method, for Linux systems without supported packages, see the
following guide:
Install on Other Linux Systems (page 9) Install MongoDB on other Linux systems from the MongoDB archives.

MongoDB Documentation, Release 2.4.9

Install MongoDB on Red Hat Enterprise, CentOS, Fedora, or Amazon Linux This tutorial outlines the steps
to install MongoDB on Red Hat Enterprise Linux, CentOS Linux, Fedora Linux and related systems. The tutorial
uses .rpm packages to install. While some of these distributions include their own MongoDB packages, the official
MongoDB packages are generally more up to date.
Packages The MongoDB downloads repository contains two packages:
mongo-10gen-server
This package contains the mongod and mongos daemons from the latest stable release and associated configuration and init scripts. Additionally, you can use this package to install daemons from a previous release
(page 4) of MongoDB.
mongo-10gen
This package contains all MongoDB tools from the latest stable release. Additionally, you can use this package
to install tools from a previous release (page 4) of MongoDB. Install this package on all production MongoDB
hosts and optionally on other systems from which you may need to administer MongoDB systems.
Install MongoDB
Configure Package Management System (YUM) Create a /etc/yum.repos.d/mongodb.repo file to hold
the following configuration information for the MongoDB repository:
Tip
For production deployments, always run MongoDB on 64-bit systems.
If you are running a 64-bit system, use the following configuration:
[mongodb]
name=MongoDB Repository
baseurl=http://downloads-distro.mongodb.org/repo/redhat/os/x86_64/
gpgcheck=0
enabled=1

If you are running a 32-bit system, which is not recommended for production deployments, use the following configuration:
[mongodb]
name=MongoDB Repository
baseurl=http://downloads-distro.mongodb.org/repo/redhat/os/i686/
gpgcheck=0
enabled=1

Install Packages Issue the following command (as root or with sudo) to install the latest stable version of MongoDB and the associated tools:
yum install mongo-10gen mongo-10gen-server

When this command completes, you have successfully installed MongoDB!

Manage Installed Versions You can use the mongo-10gen and mongo-10gen-server packages to install
previous releases of MongoDB. To install a specific release, append the version number, as in the following example:

Chapter 1. Install MongoDB

MongoDB Documentation, Release 2.4.9

yum install mongo-10gen-2.2.3 mongo-10gen-server-2.2.3

This installs the mongo-10gen and mongo-10gen-server packages with the 2.2.3 release. You can specify
any available version of MongoDB; however yum will upgrade the mongo-10gen and mongo-10gen-server
packages when a newer version becomes available. Use the following pinning procedure to prevent unintended upgrades.
To pin a package, add the following line to your /etc/yum.conf file:
exclude=mongo-10gen,mongo-10gen-server

Control Scripts

Warning: With the introduction of systemd in Fedora 15, the control scripts included in the packages available
in the MongoDB downloads repository are not compatible with Fedora systems. A correction is forthcoming,
see SERVER-7285a for more information, and in the mean time use your own control scripts or install using the
procedure outlined in Install MongoDB on Linux Systems (page 9).
a https://jira.mongodb.org/browse/SERVER-7285

The packages include various control scripts, including the init script /etc/rc.d/init.d/mongodb. These
packages configure MongoDB using the /etc/mongodb.conf file in conjunction with the control scripts.
As of version 2.4.9, there are no control scripts for mongos. mongos is only used in sharding deployments
(page 484). You can use the mongod init script to derive your own mongos control script.
Run MongoDB
Important: You must configure SELinux to allow MongoDB to start on Fedora systems. Administrators have two
options:
enable access to the relevant ports (e.g. 27017) for SELinux. See Configuration Options (page 242) for more
information on MongoDBs default ports (page 275).
disable SELinux entirely. This requires a system reboot and may have larger implications for your deployment.

Start MongoDB The MongoDB instance stores its data files in the /var/lib/mongo and its log files in
/var/log/mongo, and run using the mongod user account. If you change the user that runs the MongoDB process,
you must modify the access control rights to the /var/lib/mongo and /var/log/mongo directories.
Start the mongod process by issuing the following command (as root or with sudo):
service mongod start

You can verify that the mongod process has started successfully by checking the contents of the log file at
/var/log/mongo/mongod.log.
You may optionally ensure that MongoDB will start following a system reboot by issuing the following command
(with root privileges:)
chkconfig mongod on

Stop MongoDB Stop the mongod process by issuing the following command (as root or with sudo):
service mongod stop

1.1. Installation Guides

MongoDB Documentation, Release 2.4.9

Restart MongoDB You can restart the mongod process by issuing the following command (as root or with sudo):
service mongod restart

Follow the state of this process by watching the output in the /var/log/mongo/mongod.log file to watch for
errors or important messages from the server.
Install MongoDB on Ubuntu This tutorial outlines the steps to install MongoDB on Ubuntu Linux systems. The
tutorial uses .deb packages to install. Although Ubuntu include its own MongoDB packages, the official MongoDB
packages are generally more up to date.
Note: If you use an older Ubuntu that does not use Upstart, (i.e. any version before 9.10 Karmic) please follow the
instructions on the Install MongoDB on Debian (page 7) tutorial.

Package Options The MongoDB downloads repository provides the mongodb-10gen package, which contains
the latest stable release. Additionally you can install previous releases (page 6) of MongoDB.
You cannot install this package concurrently with the mongodb, mongodb-server, or mongodb-clients packages provided by Ubuntu.
Install MongoDB
Configure Package Management System (APT) The Ubuntu package management tool (i.e. dpkg and apt)
ensure package consistency and authenticity by requiring that distributors sign packages with GPG keys. Issue the
following command to import the MongoDB public GPG Key1 :
sudo apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv 7F0CEB10

Create a /etc/apt/sources.list.d/mongodb.list file using the following command.

echo 'deb http://downloads-distro.mongodb.org/repo/ubuntu-upstart dist 10gen' | sudo tee /etc/apt/sou

Now issue the following command to reload your repository:

sudo apt-get update

Install Packages Issue the following command to install the latest stable version of MongoDB:
sudo apt-get install mongodb-10gen

When this command completes, you have successfully installed MongoDB! Continue for configuration and start-up
suggestions.
Manage Installed Versions You can use the mongodb-10gen package to install previous versions of MongoDB.
To install a specific release, append the version number to the package name, as in the following example:
apt-get install mongodb-10gen=2.2.3

This will install the 2.2.3 release of MongoDB. You can specify any available version of MongoDB; however
apt-get will upgrade the mongodb-10gen package when a newer version becomes available. Use the following
pinning procedure to prevent unintended upgrades.
1 http://docs.mongodb.org/10gen-gpg-key.asc

Chapter 1. Install MongoDB

MongoDB Documentation, Release 2.4.9

To pin a package, issue the following command at the system prompt to pin the version of MongoDB at the currently
installed version:
echo "mongodb-10gen hold" | sudo dpkg --set-selections

Control Scripts The packages include various control scripts,

including the init script
/etc/rc.d/init.d/mongodb. These packages configure MongoDB using the /etc/mongodb.conf
file in conjunction with the control scripts.
As of version 2.4.9, there are no control scripts for mongos. mongos is only used in sharding deployments
(page 484). You can use the mongod init script to derive your own mongos control script.
Run MongoDB The MongoDB instance stores its data files in the /var/lib/mongo and its log files in
/var/log/mongo, and run using the mongod user account. If you change the user that runs the MongoDB process,
you must modify the access control rights to the /var/lib/mongo and /var/log/mongo directories.
Start MongoDB You can start the mongod process by issuing the following command:
sudo service mongodb start

You can verify that mongod has started successfully by checking the contents of the log file at
/var/log/mongodb/mongodb.log.
Stop MongoDB As needed, you may stop the mongod process by issuing the following command:
sudo service mongodb stop

Restart MongoDB You may restart the mongod process by issuing the following command:
sudo service mongodb restart

Install MongoDB on Debian This tutorial outlines the steps to install MongoDB on Debian systems. The tutorial
uses .deb packages to install. While some Debian distributions include their own MongoDB packages, the official
MongoDB packages are generally more up to date.
Note: This tutorial applies to both Debian systems and versions of Ubuntu Linux prior to 9.10 Karmic which do
not use Upstart. Other Ubuntu users will want to follow the Install MongoDB on Ubuntu (page 6) tutorial.

Package Options The downloads repository provides the mongodb-10gen package, which contains the latest
stable release. Additionally you can install previous releases (page 8) of MongoDB.
You cannot install this package concurrently with the mongodb, mongodb-server, or mongodb-clients packages that your release of Debian may include.
Install MongoDB
Configure Package Management System (APT) The Debian package management tools (i.e. dpkg and apt)
ensure package consistency and authenticity by requiring that distributors sign packages with GPG keys.

1.1. Installation Guides

MongoDB Documentation, Release 2.4.9

Step 1: Import MongoDB PGP key. Issue the following command to add the MongoDB public GPG Key2 to the
system key ring.
sudo apt-key adv --keyserver keyserver.ubuntu.com --recv 7F0CEB10

Step 2: Create a sources.list file for MongoDB. Create a /etc/apt/sources.list.d/mongodb.list

file

echo 'deb http://downloads-distro.mongodb.org/repo/debian-sysvinit dist 10gen' | sudo tee /etc/apt/so

Step 3: Reload local package database. Issue the following command to reload the local package database:
sudo apt-get update

Install Packages Issue the following command to install the latest stable version of MongoDB:
sudo apt-get install mongodb-10gen

When this command completes, you have successfully installed MongoDB!

Manage Installed Versions You can use the mongodb-10gen package to install previous versions of MongoDB.
To install a specific release, append the version number to the package name, as in the following example:
apt-get install mongodb-10gen=2.2.3

This will install the 2.2.3 release of MongoDB. You can specify any available version of MongoDB; however
apt-get will upgrade the mongodb-10gen package when a newer version becomes available. Use the following
pinning procedure to prevent unintended upgrades.
To pin a package, issue the following command at the system prompt to pin the version of MongoDB at the currently
installed version:
echo "mongodb-10gen hold" | sudo dpkg --set-selections

Control Scripts The packages include various control scripts,

including the init script
/etc/rc.d/init.d/mongodb. These packages configure MongoDB using the /etc/mongodb.conf
file in conjunction with the control scripts.
As of version 2.4.9, there are no control scripts for mongos. mongos is only used in sharding deployments
(page 484). You can use the mongod init script to derive your own mongos control script.
Run MongoDB The MongoDB instance stores its data files in the /var/lib/mongo and its log files in
/var/log/mongo, and run using the mongod user account. If you change the user that runs the MongoDB process,
you must modify the access control rights to the /var/lib/mongo and /var/log/mongo directories.
Start MongoDB Issue the following command to start mongod:
sudo /etc/init.d/mongodb start

You can verify that mongod has started successfully by checking the contents of the log file at
/var/log/mongodb/mongodb.log.
2 http://docs.mongodb.org/10gen-gpg-key.asc

Chapter 1. Install MongoDB

MongoDB Documentation, Release 2.4.9

Stop MongoDB Issue the following command to stop mongod:

sudo /etc/init.d/mongodb stop

Restart MongoDB Issue the following command to restart mongod:

sudo /etc/init.d/mongodb restart

Install MongoDB on Linux Systems Compiled versions of MongoDB for Linux provide a simple option for installing MongoDB for other Linux systems without supported packages.
Installation Process MongoDB provides archives for both 64-bit and 32-bit Linux. Follow the installation procedure
appropriate for your system.
Install for 64-bit Linux
Step 1: Download the Latest Release In a system shell, download the latest release for 64-bit Linux.
curl -O http://downloads.mongodb.org/linux/mongodb-linux-x86_64-2.4.9.tgz

You may optionally specify a different version to download.

Step 2: Extract MongoDB From Archive Extract the files from the downloaded archive.
tar -zxvf mongodb-linux-x86_64-2.4.9.tgz

Step 3: Optional. Copy MongoDB to Target Directory Copy the extracted folder into another location, such as
mongodb.
mkdir -p mongodb
cp -R -n mongodb-linux-x86_64-2.4.9/ mongodb

Step 4: Optional. Configure Search Path To ensure that the downloaded binaries are in your PATH, you can
modify your PATH and/or create symbolic links to the MongoDB binaries in your /usr/local/bin directory
(/usr/local/bin is already in your PATH). You can find the MongoDB binaries in the bin/ directory within the
archive.
Install for 32-bit Linux
Step 1: Download the Latest Release In a system shell, download the latest release for 32-bit Linux.
curl -O http://downloads.mongodb.org/linux/mongodb-linux-i686-2.4.9.tgz

You may optionally specify a different version to download.

Step 2: Extract MongoDB From Archive Extract the files from the downloaded archive.

1.1. Installation Guides

MongoDB Documentation, Release 2.4.9

tar -zxvf mongodb-linux-i686-2.4.9.tgz

Step 3: Optional. Copy MongoDB to Target Directory Copy the extracted folder into another location, such as
mongodb.
mkdir -p mongodb
cp -R -n mongodb-linux-i686-2.4.9/ mongodb

Step 4: Optional. Configure Search Path To ensure that the downloaded binaries are in your PATH, you can
modify your PATH and/or create symbolic links to the MongoDB binaries in your /usr/local/bin directory
(/usr/local/bin is already in your PATH). You can find the MongoDB binaries in the bin/ directory within the
archive.
Run MongoDB
Set Up the Data Directory Before you start mongod for the first time, you will need to create the data directory
(i.e. dbpath). By default, mongod writes data to the /data/db directory.
Step 1: Create dbpath To create the default dbpath directory, use the following command:
mkdir -p /data/db

Step 2: Set dbpath Permissions Ensure that the user that runs the mongod process has read and write permissions
to this directory. For example, if you will run the mongod process, change the owner of the /data/db directory:
chown mongodb /data/db

You must create the mongodb user separately.

You can specify an alternate path for data files using the --dbpath option to mongod. If you use an alternate
location for your data directory, ensure that this user can write to the alternate data directory.
Start MongoDB To start mongod, run the executable mongod at the system prompt.
For example, if your PATH includes the location of the mongod binary, enter mongod at the system prompt.
mongod

If your PATH does not include the location of the mongod binary, enter the full path to the mongod binary.
Starting mongod without any arguments starts a MongoDB instance that writes data to the /data/db directory. To
specify an alternate data directory, start mongod with the --dbpath option:
mongod --dbpath <some alternate directory>

Whether using the default /data/db or an alternate directory, ensure that the user account running mongod has read
and write permissions to the directory.
Stop MongoDB To stop the mongod instance, press Control+C in the terminal where the mongod instance is
running.

10

Chapter 1. Install MongoDB

MongoDB Documentation, Release 2.4.9

1.1.2 OS X
Install MongoDB on OS X (page 11)
Install MongoDB on OS X
Platform Support
Starting in version 2.4, MongoDB only supports OS X versions 10.6 (Snow Leopard) on Intel x86-64 and later.
MongoDB is available through the popular OS X package manager Homebrew3 or through the MongoDB Download
site.
Install MongoDB with Homebrew

Homebrew4 5 installs binary packages based on published formulae. The following commands will update brew to
the latest packages and install MongoDB.
In a terminal shell, use the following sequence of commands to updatebrew to the latest packages and install
MongoDB:
brew update
brew install mongodb

Later, if you need to upgrade MongoDB, run the following sequence of commands to update the MongoDB installation
on your system:
brew update
brew upgrade mongodb

Optionally, you can choose to build MongoDB from source. Use the following command to build MongoDB with
SSL support:
brew install mongodb --with-openssl

You can also install the latest development release of MongoDB for testing and development with the following
command:
brew install mongodb --devel

Manual Installation

Step 1: Download the Latest Release In a system shell, download the latest release for 64-bit OS X.
curl -O http://downloads.mongodb.org/osx/mongodb-osx-x86_64-2.4.9.tgz

You may optionally specify a different version to download.

3 http://brew.sh/
4 http://brew.sh/
5

Homebrew requires some initial setup and configuration. This configuration is beyond the scope of this document.

1.1. Installation Guides

11

MongoDB Documentation, Release 2.4.9

Step 2: Extract MongoDB From Archive Extract the files from the downloaded archive.
tar -zxvf mongodb-osx-x86_64-2.4.9.tgz

Step 3: Optional. Copy MongoDB to Target Directory Copy the extracted folder into another location, such as
mongodb.
mkdir -p mongodb
cp -R -n mongodb-osx-x86_64-2.4.9/ mongodb

Step 4: Optional. Configure Search Path To ensure that the downloaded binaries are in your PATH, you can
modify your PATH and/or create symbolic links to the MongoDB binaries in your /usr/local/bin directory
(/usr/local/bin is already in your PATH). You can find the MongoDB binaries in the bin/ directory within the
archive.
Run MongoDB

Set Up the Data Directory Before you start mongod for the first time, you will need to create the data directory.
By default, mongod writes data to the /data/db/ directory.
Step 1: Create dbpath To create the default dbpath directory, use the following command:
mkdir -p /data/db

Step 2: Set dbpath Permissions Ensure that the user that runs the mongod process has read and write permissions
to this directory. For example, if you will run the mongod process, change the owner of the /data/db directory:
chown `id -u` /data/db

You must create the mongodb user separately.

You can specify an alternate path for data files using the --dbpath option to mongod. If you use an alternate
location for your data directory, ensure that the alternate directory has the appropriate permissions.
Start MongoDB To start mongod, run the executable mongod at the system prompt.
For example, if your PATH includes the location of the mongod binary, enter mongod at the system prompt.
mongod

If your PATH does not include the location of the mongod binary, enter the full path to the mongod binary.
The previous command starts a mongod instance that writes data to the /data/db/ directory. To specify an alternate
data directory, start mongod with the --dbpath option:
mongod --dbpath <some alternate directory>

Whether using the default /data/db/ or an alternate directory, ensure that the user account running mongod has
read and write permissions to the directory.
Stop MongoDB To stop the mongod instance, press Control+C in the terminal where the mongod instance is
running.

12

Chapter 1. Install MongoDB

MongoDB Documentation, Release 2.4.9

1.1.3 Windows
Install MongoDB on Windows (page 13)
Install MongoDB on Windows
Platform Support
Starting in version 2.2, MongoDB does not support Windows XP. Please use a more recent version of Windows to use
more recent releases of MongoDB.
Important: If you are running any edition of Windows Server 2008 R2 or Windows 7, please install a hotfix to
resolve an issue with memory mapped files on Windows6 .

Download MongoDB for Windows

There are three builds of MongoDB for Windows:

MongoDB for Windows Server 2008 R2 edition (i.e. 2008R2) only runs on Windows Server 2008 R2, Windows
7 64-bit, and newer versions of Windows. This build takes advantage of recent enhancements to the Windows
Platform and cannot operate on older versions of Windows.
MongoDB for Windows 64-bit runs on any 64-bit version of Windows newer than Windows XP, including
Windows Server 2008 R2 and Windows 7 64-bit.
MongoDB for Windows 32-bit runs on any 32-bit version of Windows newer than Windows XP. 32-bit versions
of MongoDB are only intended for older systems and for use in testing and development systems. 32-bit versions
of MongoDB only support databases smaller than 2GB.
Tip
To find which version of Windows you are running, enter the following command in the Command Prompt:
wmic os get osarchitecture

1. Download the latest production release of MongoDB from the MongoDB downloads page7 . Ensure you download the correct version of MongoDB for your Windows system. The 64-bit versions of MongoDB will not
work with 32-bit Windows.
2. Extract the downloaded archive.
(a) In Windows Explorer, find the MongoDB download file, typically in the default Downloads directory.
(b) Extract the archive to C:\ by right clicking on the archive and selecting Extract All and browsing to C:\.
3. Optional. Move the MongoDB directory to another location.
C:\mongodb directory:

For example, to move the directory to

(a) Go Start Menu > All Programs > Accessories.

(b) Right click Command Prompt, and select Run as Administrator from the popup menu.
(c) In the Command Prompt, issue the following commands:
6 http://support.microsoft.com/kb/2731284
7 http://www.mongodb.org/downloads

1.1. Installation Guides

13

MongoDB Documentation, Release 2.4.9

cd \
move C:\mongodb-win32-* C:\mongodb

Note: MongoDB is self-contained and does not have any other system dependencies. You can run MongoDB from
any folder you choose. You may install MongoDB in any directory (e.g. D:\test\mongodb)

Run MongoDB

Set Up the Data Directory MongoDB requires a data folder to store its files. The default location for the MongoDB
data directory is C:\data\db. Create this folder using the Command Prompt. Go to the C:\ directory and issue the
following command sequence:
md data
md data\db

You can specify an alternate path for data files using the --dbpath option to mongod.exe.
Start MongoDB To start MongoDB, execute from the Command Prompt:
C:\mongodb\bin\mongod.exe

This will start the main MongoDB database process. The waiting for connections message in the console
output indicates that the mongod.exe process is running successfully.
Note: Depending on the security level of your system, Windows will issue a Security Alert dialog box about blocking
some features of C:\\mongodb\bin\mongod.exe from communicating on networks. All users should select
Private Networks, such as my home or work network and click Allow access. For additional
information on security and MongoDB, please read the Security Concepts (page 239) page.
Warning: Do not allow mongod.exe to be accessible to public networks without running in Secure Mode (i.e.
auth.) MongoDB is designed to be run in trusted environments and the database does not enable authentication
or Secure Mode by default.
You may specify an alternate path for \data\db with the dbpath setting for mongod.exe, as in the following
example:
C:\mongodb\bin\mongod.exe --dbpath d:\test\mongodb\data

If your path includes spaces, enclose the entire path in double quotations, for example:
C:\mongodb\bin\mongod.exe --dbpath "d:\test\mongo db data"

Connect to MongoDB Connect to MongoDB using the mongo.exe shell. Open another Command Prompt and
issue the following command:
C:\mongodb\bin\mongo.exe

Note: Executing the command start C:\mongodb\bin\mongo.exe will automatically start the mongo.exe
shell in a separate Command Prompt window.

14

Chapter 1. Install MongoDB

MongoDB Documentation, Release 2.4.9

The mongo.exe shell will connect to mongod.exe running on the localhost interface and port 27017 by default.
At the mongo.exe prompt, issue the following two commands to insert a record in the test collection of the default
test database and then retrieve that record:
db.test.save( { a: 1 } )
db.test.find()

See also:
mongo and http://docs.mongodb.org/manualreference/method. If you want to develop applications
using .NET, see the documentation of C# and MongoDB8 for more information.
MongoDB as a Windows Service

New in version 2.0.

You can set up MongoDB as a Windows Service so that the database will start automatically following each reboot
cycle.
Note: mongod.exe added support for running as a Windows service in version 2.0, and mongos.exe added
support for running as a Windows Service in version 2.1.1.

Configure the System The following steps, although optional, are good practice.
You should specify two options when running MongoDB as a Windows Service: a path for the log output (i.e.
logpath) and a configuration file.
1. Optional. Create a specific directory for MongoDB log files:
md C:\mongodb\log

2. Optional. Create a configuration file for the logpath option for MongoDB in the Command Prompt by issuing
this command:
echo logpath=C:\mongodb\log\mongo.log > C:\mongodb\mongod.cfg

Note: Consider setting the logappend option. If you do not, mongod.exe will delete the contents of the existing
log file when starting.
Changed in version 2.2: The default logpath and logappend behavior changed in the 2.2 release.

Install and Run the MongoDB Service Run all of the following commands in Command Prompt with Administrative Privileges:
1. To install the MongoDB service:
C:\mongodb\bin\mongod.exe --config C:\mongodb\mongod.cfg --install

Modify the path to the mongod.cfg file as needed. For the --install option to succeed, you must specify
a logpath setting or the --logpath run-time option.
2. To run the MongoDB service:
net start MongoDB
8 http://docs.mongodb.org/ecosystem/drivers/csharp

1.1. Installation Guides

15

MongoDB Documentation, Release 2.4.9

If you wish to use an alternate path for your dbpath specify it in the config file (e.g. C:\mongodb\mongod.cfg)
on that you specified in the --install operation. You may also specify --dbpath on the command line; however,
always prefer the configuration file.
If you have not set up the data directory, set up the data directory (page 14) where MongoDB will store its data files.
If the dbpath directory does not exist, mongod.exe will not be able to start. The default value for dbpath is
\data\db.
Stop or Remove the MongoDB Service To stop the MongoDB service:
net stop MongoDB

To remove the MongoDB service:

C:\mongodb\bin\mongod.exe --remove

1.1.4 MongoDB Enterprise

Install MongoDB Enterprise (page 16)
Install MongoDB Enterprise
New in version 2.2.
MongoDB Enterprise9 is available on four platforms and contains support for several features related to security and
monitoring.
Required Packages

Changed in version 2.4.4: MongoDB Enterprise uses Cyrus SASL instead of GNU SASL. Earlier 2.4 Enterprise
versions use GNU SASL (libgsasl) instead. For required packages for the earlier 2.4 versions, see Earlier 2.4
Versions (page 17).
To use MongoDB Enterprise, you must install several prerequisites. The names of the packages vary by distribution
and are as follows:
Debian or Ubuntu 12.04 require:
libssl0.9.8, snmp, snmpd, cyrus-sasl2-dbg,
cyrus-sasl2-mit-dbg,
libsasl2-2,
libsasl2-dev,
libsasl2-modules,
and
libsasl2-modules-gssapi-mit. Issue a command such as the following to install these packages:

sudo apt-get install libssl0.9.8 snmp snmpd cyrus-sasl2-dbg cyrus-sasl2-mit-dbg libsasl2-2 libsa

CentOS and Red Hat Enterprise Linux 6.x and 5.x, as well as Amazon Linux AMI require:
net-snmp, net-snmp-libs, openssl, net-snmp-utils, cyrus-sasl, cyrus-sasl-lib,
cyrus-sasl-devel, and cyrus-sasl-gssapi. Issue a command such as the following to install these
packages:

sudo yum install openssl net-snmp net-snmp-libs net-snmp-utils cyrus-sasl cyrus-sasl-lib cyrus-s

SUSE Enterprise Linux requires libopenssl0_9_8, libsnmp15, slessp1-libsnmp15,

snmp-mibs, cyrus-sasl, cyrus-sasl-devel, and cyrus-sasl-gssapi. Issue a command
such as the following to install these packages:
9 http://www.mongodb.com/products/mongodb-enterprise

16

Chapter 1. Install MongoDB

MongoDB Documentation, Release 2.4.9

sudo zypper install libopenssl0_9_8 libsnmp15 slessp1-libsnmp15 snmp-mibs cyrus-sasl cyrus-sasl-

Earlier 2.4 Versions Before version 2.4.4, the 2.4 versions of MongoDB Enterprise use libgsasl10 . The required
packages for the different distributions are as follows:
Ubuntu 12.04 requires libssl0.9.8, libgsasl, snmp, and snmpd. Issue a command such as the following to install these packages:
sudo apt-get install libssl0.9.8 libgsasl7 snmp snmpd

Red Hat Enterprise Linux 6.x series and Amazon Linux AMI require openssl, libgsasl7, net-snmp,
net-snmp-libs, and net-snmp-utils. To download libgsasl you must enable the EPEL repository
by issuing the following sequence of commands to add and update the system repositories:
sudo rpm -ivh http://download.fedoraproject.org/pub/epel/6/x86_64/epel-release-6-8.noarch.rpm
sudo yum update -y

When you have installed and updated the EPEL repositories, issue the following install these packages:
sudo yum install openssl net-snmp net-snmp-libs net-snmp-utils libgsasl

SUSE Enterprise Linux requires libopenssl0_9_8, libsnmp15, slessp1-libsnmp15, and

snmp-mibs. Issue a command such as the following to install these packages:
sudo zypper install libopenssl0_9_8 libsnmp15 slessp1-libsnmp15 snmp-mibs

Note: Before 2.4.4, MongoDB Enterprise 2.4 for SUSE requires libgsasl11 which is not available in the default
repositories for SUSE.

Install MongoDB Enterprise Binaries

When you have installed the required packages, and downloaded the Enterprise packages12 you can install the packages
using the same procedure as a standard installation of MongoDB on Linux Systems (page 9).
Note: .deb and .rpm packages for Enterprise releases are available for some platforms. You can use these to install
MongoDB directly using the dpkg and rpm utilities.
Use the sequence of commands below to download and extract MongoDB Enterprise packages appropriate for your
distribution:
Ubuntu 12.04
curl -O http://downloads.10gen.com/linux/mongodb-linux-x86_64-subscription-ubuntu1204-2.4.9.tgz
tar -zxvf mongodb-linux-x86_64-subscription-ubuntu1204-2.4.9.tgz
cp -R -n mongodb-linux-x86_64-subscription-ubuntu1204-2.4.9/ mongodb

Red Hat Enterprise Linux 6.x

10 http://www.gnu.org/software/gsasl/
11 http://www.gnu.org/software/gsasl/
12 http://www.mongodb.com/products/mongodb-enterprise

1.1. Installation Guides

17

MongoDB Documentation, Release 2.4.9

curl -O http://downloads.10gen.com/linux/mongodb-linux-x86_64-subscription-rhel62-2.4.9.tgz
tar -zxvf mongodb-linux-x86_64-subscription-rhel62-2.4.9.tgz
cp -R -n mongodb-linux-x86_64-subscription-rhel62-2.4.9/ mongodb

Amazon Linux AMI

curl -O http://downloads.10gen.com/linux/mongodb-linux-x86_64-subscription-amzn64-2.4.9.tgz
tar -zxvf mongodb-linux-x86_64-subscription-amzn64-2.4.9.tgz
cp -R -n mongodb-linux-x86_64-subscription-amzn64-2.4.9/ mongodb

SUSE Enterprise Linux

curl -O http://downloads.10gen.com/linux/mongodb-linux-x86_64-subscription-suse11-2.4.9.tgz
tar -zxvf mongodb-linux-x86_64-subscription-suse11-2.4.9.tgz
cp -R -n mongodb-linux-x86_64-subscription-suse11-2.4.9/ mongodb

Running and Using MongoDB

Note: The Enterprise packages currently include an example SNMP configuration file named mongod.conf. This
file is not a MongoDB configuration file.
Before you start mongod for the first time, you will need to create the data directory (i.e. dbpath). By default,
mongod writes data to the /data/db directory.
Step 1: Create dbpath To create the default dbpath directory, use the following command:
mkdir -p /data/db

Step 2: Set dbpath Permissions Ensure that the user that runs the mongod process has read and write permissions
to this directory. For example, if you will run the mongod process, change the owner of the /data/db directory:
chown mongodb /data/db

You must create the mongodb user separately.

You can specify an alternate path for data files using the --dbpath option to mongod. If you use an alternate
location for your data directory, ensure that this user can write to the alternate data directory.
Start MongoDB To start mongod, run the executable mongod at the system prompt.
For example, if your PATH includes the location of the mongod binary, enter mongod at the system prompt.
mongod

If your PATH does not include the location of the mongod binary, enter the full path to the mongod binary.
Starting mongod without any arguments starts a MongoDB instance that writes data to the /data/db directory. To
specify an alternate data directory, start mongod with the --dbpath option:
mongod --dbpath <some alternate directory>

Whether using the default /data/db or an alternate directory, ensure that the user account running mongod has read
and write permissions to the directory.

18

Chapter 1. Install MongoDB

MongoDB Documentation, Release 2.4.9

Stop MongoDB To stop the mongod instance, press Control+C in the terminal where the mongod instance is
running.
Further Reading

As you begin to use MongoDB, consider the Getting Started with MongoDB (page 19) and MongoDB Tutorials
(page 186) resources. To read about features only available in MongoDB Enterprise, consider: Monitor MongoDB
with SNMP (page 177) and Deploy MongoDB with Kerberos Authentication (page 261).

1.2 First Steps with MongoDB

After you have installed MongoDB, consider the following documents as you begin to learn about MongoDB:
Getting Started with MongoDB (page 19) An introduction to the basic operation and use of MongoDB.
Generate Test Data (page 23) To support initial exploration, generate test data to facilitate testing.

1.2.1 Getting Started with MongoDB

This tutorial provides an introduction to basic database operations using the mongo shell. mongo is a part of the
standard MongoDB distribution and provides a full JavaScript environment with a complete access to the JavaScript
language and all standard functions as well as a full database interface for MongoDB. See the mongo JavaScript API13
documentation and the mongo shell JavaScript Method Reference.
The tutorial assumes that youre running MongoDB on a Linux or OS X operating system and that you have a running
database server; MongoDB does support Windows and provides a Windows distribution with identical operation.
For instructions on installing MongoDB and starting the database server, see the appropriate installation (page 3)
document.
Connect to a Database
In this section, you connect to the database server, which runs as mongod, and begin using the mongo shell to select
a logical database within the database instance and access the help text in the mongo shell.
Connect to a mongod

From a system prompt, start mongo by issuing the mongo command, as follows:
mongo

By default, mongo looks for a database server listening on port 27017 on the localhost interface. To connect to
a server on a different port or interface, use the --port and --host options.
Select a Database

After starting the mongo shell your session will use the test database by default. At any time, issue the following
operation at the mongo to report the name of the current database:
13 http://api.mongodb.org/js

1.2. First Steps with MongoDB

19

MongoDB Documentation, Release 2.4.9

db

1. From the mongo shell, display the list of databases, with the following operation:
show dbs

2. Switch to a new database named mydb, with the following operation:

use mydb

3. Confirm that your session has the mydb database as context, by checking the value of the db object, which
returns the name of the current database, as follows:
db

At this point, if you issue the show dbs operation again, it will not include the mydb database. MongoDB
will not permanently create a database until you insert data into that database. The Create a Collection and
Insert Documents (page 20) section describes the process for inserting data.
New in version 2.4: show databases also returns a list of databases.
Display mongo Help

At any point, you can access help for the mongo shell using the following operation:
help

Furthermore, you can append the .help() method to some JavaScript methods, any cursor object, as well as the db
and db.collection objects to return additional help information.
Create a Collection and Insert Documents
In this section, you insert documents into a new collection named testData within the new database named mydb.
MongoDB will create a collection implicitly upon its first use. You do not need to create a collection before inserting
data. Furthermore, because MongoDB uses dynamic schemas (page 556), you also need not specify the structure of
your documents before inserting them into the collection.
1. From the mongo shell, confirm you are in the mydb database by issuing the following:
db

2. If mongo does not return mydb for the previous operation, set the context to the mydb database, with the
following operation:
use mydb

3. Create two documents named j and k by using the following sequence of JavaScript operations:
j = { name : "mongo" }
k = { x : 3 }

4. Insert the j and k documents into the testData collection with the following sequence of operations:
db.testData.insert( j )
db.testData.insert( k )

When you insert the first document, the mongod will create both the mydb database and the testData
collection.

20

Chapter 1. Install MongoDB

MongoDB Documentation, Release 2.4.9

5. Confirm that the testData collection exists. Issue the following operation:
show collections

The mongo shell will return the list of the collections in the current (i.e. mydb) database. At this point, the only
collection is testData. All mongod databases also have a system.indexes (page 227) collection.
6. Confirm that the documents exist in the testData collection by issuing a query on the collection using the
find() method:
db.testData.find()

This operation returns the following results. The ObjectId (page 129) values will be unique:
{ "_id" : ObjectId("4c2209f9f3924d31102bd84a"), "name" : "mongo" }
{ "_id" : ObjectId("4c2209fef3924d31102bd84b"), "x" : 3 }

All MongoDB documents must have an _id field with a unique value. These operations do not explicitly
specify a value for the _id field, so mongo creates a unique ObjectId (page 129) value for the field before
inserting it into the collection.
Insert Documents using a For Loop or a JavaScript Function
To perform the remaining procedures in this tutorial, first add more documents to your database using one or both of
the procedures described in Generate Test Data (page 23).
Working with the Cursor
When you query a collection, MongoDB returns a cursor object that contains the results of the query. The mongo
shell then iterates over the cursor to display the results. Rather than returning all results at once, the shell iterates over
the cursor 20 times to display the first 20 results and then waits for a request to iterate over the remaining results. In
the shell, use enter it to iterate over the next set of results.
The procedures in this section show other ways to work with a cursor. For comprehensive documentation on cursors,
see crud-read-cursor.
Iterate over the Cursor with a Loop

Before using this procedure, make sure to add at least 25 documents to a collection using one of the procedures in
Generate Test Data (page 23). You can name your database and collections anything you choose, but this procedure
will assume the database named test and a collection named testData.
1. In the MongoDB JavaScript shell, query the testData collection and assign the resulting cursor object to the
c variable:
var c = db.testData.find()

2. Print the full result set by using a while loop to iterate over the c variable:
while ( c.hasNext() ) printjson( c.next() )

The hasNext() function returns true if the cursor has documents. The next() method returns the next
document. The printjson() method renders the document in a JSON-like format.
The operation displays 20 documents. For example, if the documents have a single field named x, the operation
displays the field as well as each documents ObjectId:

1.2. First Steps with MongoDB

21

MongoDB Documentation, Release 2.4.9

{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{

"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"

:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:

ObjectId("51a7dc7b2cacf40b79990be6"),
ObjectId("51a7dc7b2cacf40b79990be7"),
ObjectId("51a7dc7b2cacf40b79990be8"),
ObjectId("51a7dc7b2cacf40b79990be9"),
ObjectId("51a7dc7b2cacf40b79990bea"),
ObjectId("51a7dc7b2cacf40b79990beb"),
ObjectId("51a7dc7b2cacf40b79990bec"),
ObjectId("51a7dc7b2cacf40b79990bed"),
ObjectId("51a7dc7b2cacf40b79990bee"),
ObjectId("51a7dc7b2cacf40b79990bef"),
ObjectId("51a7dc7b2cacf40b79990bf0"),
ObjectId("51a7dc7b2cacf40b79990bf1"),
ObjectId("51a7dc7b2cacf40b79990bf2"),
ObjectId("51a7dc7b2cacf40b79990bf3"),
ObjectId("51a7dc7b2cacf40b79990bf4"),
ObjectId("51a7dc7b2cacf40b79990bf5"),
ObjectId("51a7dc7b2cacf40b79990bf6"),
ObjectId("51a7dc7b2cacf40b79990bf7"),
ObjectId("51a7dc7b2cacf40b79990bf8"),
ObjectId("51a7dc7b2cacf40b79990bf9"),

"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"

:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:

1 }
2 }
3 }
4 }
5 }
6 }
7 }
8 }
9 }
10 }
11 }
12 }
13 }
14 }
15 }
16 }
17 }
18 }
19 }
20 }

Use Array Operations with the Cursor

The following procedure lets you manipulate a cursor object as if it were an array:
1. In the mongo shell, query the testData collection and assign the resulting cursor object to the c variable:
var c = db.testData.find()

2. To find the document at the array index 4, use the following operation:
printjson( c [ 4 ] )

MongoDB returns the following:

{ "_id" : ObjectId("51a7dc7b2cacf40b79990bea"), "x" : 5 }

When you access documents in a cursor using the array index notation, mongo first calls the
cursor.toArray() method and loads into RAM all documents returned by the cursor. The index is then
applied to the resulting array. This operation iterates the cursor completely and exhausts the cursor.
For very large result sets, mongo may run out of available memory.
For more information on the cursor, see crud-read-cursor.
Query for Specific Documents

MongoDB has a rich query system that allows you to select and filter the documents in a collection along specific
fields and values. See Query Documents (page 60) and Read Operations (page 31) for a full account of queries in
MongoDB.
In this procedure, you query for specific documents in the testData collection by passing a query document as a
parameter to the find() method. A query document specifies the criteria the query must match to return a document.
In the mongo shell, query for all documents where the x field has a value of 18 by passing the { x :
document as a parameter to the find() method:

22

18 } query

Chapter 1. Install MongoDB

MongoDB Documentation, Release 2.4.9

db.testData.find( { x : 18 } )

MongoDB returns one document that fits this criteria:

{ "_id" : ObjectId("51a7dc7b2cacf40b79990bf7"), "x" : 18 }

Return a Single Document from a Collection

With the findOne() method you can return a single document from a MongoDB collection. The findOne()
method takes the same parameters as find(), but returns a document rather than a cursor.
To retrieve one document from the testData collection, issue the following command:
db.testData.findOne()

For more information on querying for documents, see the Query Documents (page 60) and Read Operations (page 31)
documentation.
Limit the Number of Documents in the Result Set

To increase performance, you can constrain the size of the result by limiting the amount of data your application must
receive over the network.
To specify the maximum number of documents in the result set, call the limit() method on a cursor, as in the
following command:
db.testData.find().limit(3)

MongoDB will return the following result, with different ObjectId (page 129) values:
{ "_id" : ObjectId("51a7dc7b2cacf40b79990be6"), "x" : 1 }
{ "_id" : ObjectId("51a7dc7b2cacf40b79990be7"), "x" : 2 }
{ "_id" : ObjectId("51a7dc7b2cacf40b79990be8"), "x" : 3 }

Next Steps with MongoDB

For more information on manipulating the documents in a database as you continue to learn MongoDB, consider the
following resources:
MongoDB CRUD Operations (page 27)
SQL to MongoDB Mapping Chart (page 85)
MongoDB Drivers and Client Libraries (page 95)

1.2.2 Generate Test Data

This tutorial describes how to quickly generate test data as you need to test basic MongoDB operations.
Insert Multiple Documents Using a For Loop
You can add documents to a new or existing collection by using a JavaScript for loop run from the mongo shell.

1.2. First Steps with MongoDB

23

MongoDB Documentation, Release 2.4.9

1. From the mongo shell, insert new documents into the testData collection using the following for loop. If
the testData collection does not exist, MongoDB creates the collection implicitly.
for (var i = 1; i <= 25; i++) db.testData.insert( { x : i } )

2. Use find() to query the collection:

db.testData.find()

The mongo shell displays the first 20 documents in the collection. Your ObjectId (page 129) values will be
different:
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{
{

"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"
"_id"

:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:

ObjectId("51a7dc7b2cacf40b79990be6"),
ObjectId("51a7dc7b2cacf40b79990be7"),
ObjectId("51a7dc7b2cacf40b79990be8"),
ObjectId("51a7dc7b2cacf40b79990be9"),
ObjectId("51a7dc7b2cacf40b79990bea"),
ObjectId("51a7dc7b2cacf40b79990beb"),
ObjectId("51a7dc7b2cacf40b79990bec"),
ObjectId("51a7dc7b2cacf40b79990bed"),
ObjectId("51a7dc7b2cacf40b79990bee"),
ObjectId("51a7dc7b2cacf40b79990bef"),
ObjectId("51a7dc7b2cacf40b79990bf0"),
ObjectId("51a7dc7b2cacf40b79990bf1"),
ObjectId("51a7dc7b2cacf40b79990bf2"),
ObjectId("51a7dc7b2cacf40b79990bf3"),
ObjectId("51a7dc7b2cacf40b79990bf4"),
ObjectId("51a7dc7b2cacf40b79990bf5"),
ObjectId("51a7dc7b2cacf40b79990bf6"),
ObjectId("51a7dc7b2cacf40b79990bf7"),
ObjectId("51a7dc7b2cacf40b79990bf8"),
ObjectId("51a7dc7b2cacf40b79990bf9"),

"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"
"x"

:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:
:

1 }
2 }
3 }
4 }
5 }
6 }
7 }
8 }
9 }
10 }
11 }
12 }
13 }
14 }
15 }
16 }
17 }
18 }
19 }
20 }

1. The find() returns a cursor. To iterate the cursor and return more documents use the it operation in the
mongo shell. The mongo shell will exhaust the cursor, and return the following documents:
{
{
{
{
{

"_id"
"_id"
"_id"
"_id"
"_id"

:
:
:
:
:

ObjectId("51a7dce92cacf40b79990bfc"),
ObjectId("51a7dce92cacf40b79990bfd"),
ObjectId("51a7dce92cacf40b79990bfe"),
ObjectId("51a7dce92cacf40b79990bff"),
ObjectId("51a7dce92cacf40b79990c00"),

"x"
"x"
"x"
"x"
"x"

:
:
:
:
:

21
22
23
24
25

}
}
}
}
}

Insert Multiple Documents with a mongo Shell Function

You can create a JavaScript function in your shell session to generate the above data. The insertData() JavaScript
function, shown here, creates new data for use in testing or training by either creating a new collection or appending
data to an existing collection:
function insertData(dbName, colName, num) {
var col = db.getSiblingDB(dbName).getCollection(colName);
for (i = 0; i < num; i++) {
col.insert({x:i});
}
print(col.count());

24

Chapter 1. Install MongoDB

MongoDB Documentation, Release 2.4.9

The insertData() function takes three parameters: a database, a new or existing collection, and the number of
documents to create. The function creates documents with an x field that is set to an incremented integer, as in the
following example documents:
{ "_id" : ObjectId("51a4da9b292904caffcff6eb"), "x" : 0 }
{ "_id" : ObjectId("51a4da9b292904caffcff6ec"), "x" : 1 }
{ "_id" : ObjectId("51a4da9b292904caffcff6ed"), "x" : 2 }

Store the function in your .mongorc.js file. The mongo shell loads the function for you every time you start a session.
Example
Specify database name, collection name, and the number of documents to insert as arguments to insertData().
insertData("test", "testData", 400)

This operation inserts 400 documents into the testData collection in the test database. If the collection and
database do not exist, MongoDB creates them implicitly before inserting documents.
See also:
MongoDB CRUD Concepts (page 29) and Data Models (page 97).

1.2. First Steps with MongoDB

25

MongoDB Documentation, Release 2.4.9

26

Chapter 1. Install MongoDB

CHAPTER 2

MongoDB CRUD Operations

MongoDB provides rich semantics for reading and manipulating data. CRUD stands for create, read, update, and
delete. These terms are the foundation for all interactions with the database.
MongoDB CRUD Introduction (page 27) An introduction to the MongoDB data model as well as queries and data
manipulations.
MongoDB CRUD Concepts (page 29) The core documentation of query and data manipulation.
MongoDB CRUD Tutorials (page 58) Examples of basic query and data modification operations.
MongoDB CRUD Reference (page 82) Reference material for the query and data manipulation interfaces.

2.1 MongoDB CRUD Introduction

MongoDB stores data in the form of documents, which are JSON-like field and value pairs. Documents are analogous
to structures in programming languages that associate keys with values, where keys may hold other pairs of keys and
values (e.g. dictionaries, hashes, maps, and associative arrays). Formally, MongoDB documents are BSON documents,
which is a binary representation of JSON with additional type information. For more information, see Documents
(page 122).

Figure 2.1: A MongoDB document.

MongoDB stores all documents in collections. A collection is a group of related documents that have a set of shared
common indexes. Collections are analogous to a table in relational databases.

27

MongoDB Documentation, Release 2.4.9

Figure 2.2: A collection of MongoDB documents.

2.1.1 Database Operations

Query
In MongoDB a query targets a specific collection of documents. Queries specify criteria, or conditions, that identify
the documents that MongoDB returns to the clients. A query may include a projection that specifies the fields from
the matching documents to return. You can optionally modify queries to impose limits, skips, and sort orders.
In the following diagram, the query process specifies a query criteria and a sort modifier:
Data Modification
Data modification refers to operations that create, update, or delete data. In MongoDB, these operations modify the
data of a single collection. For the update and delete operations, you can specify the criteria to select the documents
to update or remove.
In the following diagram, the insert operation adds a new document to the users collection.

2.1.2 Related Features

Indexes
To enhance the performance of common queries and updates, MongoDB has full support for secondary indexes. These
indexes allow applications to store a view of a portion of the collection in an efficient data structure. Most indexes store
an ordered representation of all values of a field or a group of fields. Indexes may also enforce uniqueness (page 334),
store objects in a geospatial representation (page 326), and facilitate text search (page 332).

28

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Figure 2.3: The stages of a MongoDB query with a query criteria and a sort modifier.
Read Preference
For replica sets and sharded clusters with replica set components, applications specify read preferences (page 406). A
read preference determines how the client direct read operations to the set.
Write Concern
Applications can also control the behavior of write operations using write concern (page 47). Particularly useful
for deployments with replica sets, the write concern semantics allow clients to specify the assurance that MongoDB
provides when reporting on the success of a write operation.
Aggregation
In addition to the basic queries, MongoDB provides several data aggregation features. For example, MongoDB can
return counts of the number of documents that match a query, or return the number of distinct values for a field, or
process a collection of documents using a versatile stage-based data processing pipeline or map-reduce operations.

2.2 MongoDB CRUD Concepts

The Read Operations (page 31) and Write Operations (page 42) documents introduce the behavior and operations of
read and write operations for MongoDB deployments.
Read Operations (page 31) Introduces all operations that select and return documents to clients, including the query
specifications.
Cursors (page 35) Queries return iterable objects, called cursors, that hold the full result set.
Query Optimization (page 36) Analyze and improve query performance.

2.2. MongoDB CRUD Concepts

29

MongoDB Documentation, Release 2.4.9

Figure 2.4: The stages of a MongoDB insert operation.

30

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Distributed Queries (page 38) Describes how sharded clusters and replica sets affect the performance of read
operations.
Write Operations (page 42) Introduces data create and modify operations, their behavior, and performances.
Write Concern (page 47) Describes the kind of guarantee MongoDB provides when reporting on the success
of a write operation.
Distributed Write Operations (page 51) Describes how MongoDB directs write operations on sharded clusters
and replica sets and the performance characteristics of these operations.

2.2.1 Read Operations

The following documents describe read operations:
Read Operations Overview (page 31) A high level overview of queries and projections in MongoDB, including a
discussion of syntax and behavior.
Cursors (page 35) Queries return iterable objects, called cursors, that hold the full result set.
Query Optimization (page 36) Analyze and improve query performance.
Query Plans (page 37) MongoDB executes queries using optimal plans.
Distributed Queries (page 38) Describes how sharded clusters and replica sets affect the performance of read operations.
Read Operations Overview
Read operations, or queries, retrieve data stored in the database. In MongoDB, queries select documents from a single
collection.
Queries specify criteria, or conditions, that identify the documents that MongoDB returns to the clients. A query may
include a projection that specifies the fields from the matching documents to return. The projection limits the amount
of data that MongoDB returns to the client over the network.
Query Interface

For query operations, MongoDB provides a db.collection.find() method. The method accepts both the
query criteria and projections and returns a cursor (page 35) to the matching documents. You can optionally modify
the query to impose limits, skips, and sort orders.
The following diagram highlights the components of a MongoDB query operation:

Figure 2.5: The components of a MongoDB find operation.

The next diagram shows the same query in SQL:
Example
2.2. MongoDB CRUD Concepts

31

MongoDB Documentation, Release 2.4.9

Figure 2.6: The components of a SQL SELECT statement.

db.users.find( { age: { $gt: 18 } }, { name: 1, address: 1 } ).limit(5)

This query selects the documents in the users collection that match the condition age is greater than 18. To specify
the greater than condition, query criteria uses the greater than (i.e. $gt) query selection operator. The query returns
at most 5 matching documents (or more precisely, a cursor to those documents). The matching documents will return
with only the _id, name and address fields. See Projections (page 33) for details.
See
SQL to MongoDB Mapping Chart (page 85) for additional examples of MongoDB queries and the corresponding SQL
statements.

Query Behavior

MongoDB queries exhibit the following behavior:

All queries in MongoDB address a single collection.
You can modify the query to impose limits, skips, and sort orders.
The order of documents returned by a query is not defined unless you specify a sort().
Operations that modify existing documents (page 67) (i.e. updates) use the same query syntax as queries to select
documents to update.
In aggregation (page 281) pipeline, the $match pipeline stage provides access to MongoDB queries.
MongoDB provides a db.collection.findOne() method as a special case of find() that returns a single
document.
Query Statements

Consider the following diagram of the query process that specifies a query criteria and a sort modifier:
In the diagram, the query selects documents from the users collection. Using a query selection operator
to define the conditions for matching documents, the query selects documents that have age greater than (i.e. $gt)
18. Then the sort() modifier sorts the results by age in ascending order.
For additional examples of queries, see Query Documents (page 60).

32

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Figure 2.7: The stages of a MongoDB query with a query criteria and a sort modifier.
Projections

Queries in MongoDB return all fields in all matching documents by default. To limit the amount of data that MongoDB
sends to applications, include a projection in the queries. By projecting results with a subset of fields, applications
reduce their network overhead and processing requirements.
Projections, which are the the second argument to the find() method, may either specify a list of fields to return or
list fields to exclude in the result documents.
Important:
projections.

Except for excluding the _id field in inclusive projections, you cannot mix exclusive and inclusive

Consider the following diagram of the query process that specifies a query criteria and a projection:
In the diagram, the query selects from the users collection. The criteria matches the documents that have age equal
to 18. Then the projection specifies that only the name field should return in the matching documents.
Projection Examples
Exclude One Field From a Result Set
db.records.find( { "user_id": { $lt: 42} }, { history: 0} )

This query selects a number of documents in the records collection that match the query { "user_id":
$lt: 42} }, but excludes the history field.

Return Two fields and the _id Field

db.records.find( { "user_id": { $lt: 42} }, { "name": 1, "email": 1} )

2.2. MongoDB CRUD Concepts

33

MongoDB Documentation, Release 2.4.9

Figure 2.8: The stages of a MongoDB query with a query criteria and projection. MongoDB only transmits the
projected data to the clients.
This query selects a number of documents in the records collection that match the query { "user_id": {
$lt: 42} }, but returns documents that have the _id field (implicitly included) as well as the name and email
fields.
Return Two Fields and Exclude _id
db.records.find( { "user_id": { $lt: 42} }, { "_id": 0, "name": 1 , "email": 1 } )

This query selects a number of documents in the records collection that match the query { "user_id":
$lt: 42} }, but only returns the name and email fields.

See
Limit Fields to Return from a Query (page 64) for more examples of queries with projection statements.

Projection Behavior MongoDB projections have the following properties:

In MongoDB, the _id field is always included in results unless explicitly excluded.
For fields that contain arrays, MongoDB provides the following projection operators: $elemMatch, $slice,
$.
For related projection functionality in the aggregation framework (page 281) pipeline, use the $project
pipeline stage.

34

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Cursors
In the mongo shell, the primary method for the read operation is the db.collection.find() method. This
method queries a collection and returns a cursor to the returning documents.
To access the documents, you need to iterate the cursor. However, in the mongo shell, if the returned cursor is not
assigned to a variable using the var keyword, then the cursor is automatically iterated up to 20 times 1 to print up to
the first 20 documents in the results.
For example, in the mongo shell, the following read operation queries the inventory collection for documents that
have type equal to food and automatically print up to the first 20 matching documents:
db.inventory.find( { type: 'food' } );

To manually iterate the cursor to access the documents, see Iterate a Cursor in the mongo Shell (page 65).
Cursor Behaviors

Closure of Inactive Cursors By default, the server will automatically close the cursor after 10 minutes of inactivity
or if client has exhausted the cursor. To override this behavior, you can specify the noTimeout wire protocol flag2
in your query; however, you should either close the cursor manually or exhaust the cursor. In the mongo shell, you
can set the noTimeout flag:
var myCursor = db.inventory.find().addOption(DBQuery.Option.noTimeout);

See your driver (page 95) documentation for information on setting the noTimeout flag. For the mongo shell, see
cursor.addOption() for a complete list of available cursor flags.
Cursor Isolation Because the cursor is not isolated during its lifetime, intervening write operations on a document
may result in a cursor that returns a document more than once if that document has changed. To handle this situation,
see the information on snapshot mode (page 565).
Cursor Batches The MongoDB server returns the query results in batches. Batch size will not exceed the maximum
BSON document size. For most queries, the first batch returns 101 documents or just enough documents to exceed 1
megabyte. Subsequent batch size is 4 megabytes. To override the default size of the batch, see batchSize() and
limit().
For queries that include a sort operation without an index, the server must load all the documents in memory to perform
the sort and will return all documents in the first batch.
As you iterate through the cursor and reach the end of the returned batch, if there are more results, cursor.next()
will perform a getmore operation to retrieve the next batch. To see how many documents remain in the batch
as you iterate the cursor, you can use the objsLeftInBatch() method, as in the following example:
var myCursor = db.inventory.find();
var myFirstDocument = myCursor.hasNext() ? myCursor.next() : null;
myCursor.objsLeftInBatch();
1 You can use the DBQuery.shellBatchSize to change the number of iteration from the default value 20. See Executing Queries
(page 213) for more information.
2 http://docs.mongodb.org/meta-driver/latest/legacy/mongodb-wire-protocol

2.2. MongoDB CRUD Concepts

35

MongoDB Documentation, Release 2.4.9

Cursor Information

You can use the command cursorInfo to retrieve the following information on cursors:
total number of open cursors
size of the client cursors in current use
number of timed out cursors since the last server restart
Consider the following example:
db.runCommand( { cursorInfo: 1 } )

The result from the command returns the following document:

{
"totalOpen" : <number>,
"clientCursors_size" : <number>,
"timedOut" : <number>,
"ok" : 1
}

Query Optimization
Indexes improve the efficiency of read operations by reducing the amount of data that query operations need to process.
This simplifies the work associated with fulfilling queries within MongoDB.
Create an Index to Support Read Operations

If your application queries a collection on a particular field or fields, then an index on the queried field or fields can
prevent the query from scanning the whole collection to find and return the query results. For more information about
indexes, see the complete documentation of indexes in MongoDB (page 318).
Example
An application queries the inventory collection on the type field. The value of the type field is user-driven.
var typeValue = <someUserInput>;
db.inventory.find( { type: typeValue } );

To improve the performance of this query, add an ascending, or a descending, index to the inventory collection
on the type field. 3 In the mongo shell, you can create indexes using the db.collection.ensureIndex()
method:
db.inventory.ensureIndex( { type: 1 } )

This index can prevent the above query on type from scanning the whole collection to return the results.
To analyze the performance of the query with an index, see Analyze Query Performance (page 66).
In addition to optimizing read operations, indexes can support sort operations and allow for a more efficient storage
utilization. See db.collection.ensureIndex() and Indexing Tutorials (page 339) for more information about
index creation.
3 For single-field indexes, the selection between ascending and descending order is immaterial. For compound indexes, the selection is important.
See indexing order (page 323) for more details.

36

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Query Selectivity

Some query operations are not selective. These operations cannot use indexes effectively or cannot use indexes at all.
The inequality operators $nin and $ne are not very selective, as they often match a large portion of the index. As a
result, in most cases, a $nin or $ne query with an index may perform no better than a $nin or $ne query that must
scan all documents in a collection.
Queries that specify regular expressions, with inline JavaScript regular expressions or $regex operator expressions,
cannot use an index with one exception. Queries that specify regular expression with anchors at the beginning of a
string can use an index.
Covering a Query

An index covers (page 369) a query, a covered query, when:

all the fields in the query (page 60) are part of that index, and
all the fields returned in the documents that match the query are in the same index.
For these queries, MongoDB does not need to inspect documents outside of the index. This is often more efficient
than inspecting entire documents.
Example
Given a collection inventory with the following index on the type and item fields:
{ type: 1, item: 1 }

This index will cover the following query on the type and item fields, which returns only the item field:
db.inventory.find( { type: "food", item:/^c/ },
{ item: 1, _id: 0 } )

However, the index will not cover the following query, which returns the item field and the _id field:
db.inventory.find( { type: "food", item:/^c/ },
{ item: 1 } )

See Create Indexes that Support Covered Queries (page 369) for more information on the behavior and use of covered
queries.
Query Plans
The MongoDB query optimizer processes queries and chooses the most efficient query plan for a query given the available indexes. The query system then uses this query plan each time the query runs. The query optimizer occasionally
reevaluates query plans as the content of the collection changes to ensure optimal query plans.
You can use the explain() method to view statistics about the query plan for a given query. This information can
help as you develop indexing strategies (page 368).
Query Optimization

To create a new query plan, the query optimizer:

1. runs the query against several candidate indexes in parallel.

2.2. MongoDB CRUD Concepts

37

MongoDB Documentation, Release 2.4.9

2. records the matches in a common results buffer or buffers.

If the candidate plans include only ordered query plans, there is a single common results buffer.
If the candidate plans include only unordered query plans, there is a single common results buffer.
If the candidate plans include both ordered query plans and unordered query plans, there are two common
results buffers, one for the ordered plans and the other for the unordered plans.
If an index returns a result already returned by another index, the optimizer skips the duplicate match. In the
case of the two buffers, both buffers are de-duped.
3. stops the testing of candidate plans and selects an index when one of the following events occur:
An unordered query plan has returned all the matching results; or
An ordered query plan has returned all the matching results; or
An ordered query plan has returned a threshold number of matching results:
Version 2.0: Threshold is the query batch size. The default batch size is 101.
Version 2.2: Threshold is 101.
The selected index becomes the index specified in the query plan; future iterations of this query or queries with the
same query pattern will use this index. Query pattern refers to query select conditions that differ only in the values, as
in the following two queries with the same query pattern:
db.inventory.find( { type: 'food' } )
db.inventory.find( { type: 'utensil' } )

Query Plan Revision

As collections change over time, the query optimizer deletes the query plan and re-evaluates after any of the following
events:
The collection receives 1,000 write operations.
The reIndex rebuilds the index.
You add or drop an index.
The mongod process restarts.
Distributed Queries
Read Operations to Sharded Clusters

Sharded clusters allow you to partition a data set among a cluster of mongod instances in a way that is nearly transparent to the application. For an overview of sharded clusters, see the Sharding (page 479) section of this manual.
For a sharded cluster, applications issue operations to one of the mongos instances associated with the cluster.
Read operations on sharded clusters are most efficient when directed to a specific shard. Queries to sharded collections
should include the collections shard key (page 492). When a query includes a shard key, the mongos can use cluster
metadata from the config database (page 488) to route the queries to shards.
If a query does not include the shard key, the mongos must direct the query to all shards in the cluster. These scatter
gather queries can be inefficient. On larger clusters, scatter gather queries are unfeasible for routine operations.
For more information on read operations in sharded clusters, see the Sharded Cluster Query Routing (page 496) and
Shard Keys (page 492) sections.
38

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Figure 2.9: Diagram of a sharded cluster.

2.2. MongoDB CRUD Concepts

39

MongoDB Documentation, Release 2.4.9

Figure 2.10: Read operations to a sharded cluster. Query criteria includes the shard key. The query router mongos
can target the query to the appropriate shard or shards.

40

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Figure 2.11: Read operations to a sharded cluster. Query criteria does not include the shard key. The query router
mongos must broadcast query to all shards for the collection.

2.2. MongoDB CRUD Concepts

41

MongoDB Documentation, Release 2.4.9

Read Operations to Replica Sets

Replica sets use read preferences to determine where and how to route read operations to members of the replica set.
By default, MongoDB always reads data from a replica sets primary. You can modify that behavior by changing the
read preference mode (page 476).
You can configure the read preference mode (page 476) on a per-connection or per-operation basis to allow reads from
secondaries to:
reduce latency in multi-data-center deployments,
improve read throughput by distributing high read-volumes (relative to write volume),
for backup operations, and/or
to allow reads during failover (page 397) situations.

Figure 2.12: Read operations to a replica set. Default read preference routes the read to the primary. Read preference
of nearest routes the read to the nearest member.
Read operations from secondary members of replica sets are not guaranteed to reflect the current state of the primary,
and the state of secondaries will trail the primary by some amount of time. Often, applications dont rely on this kind
of strict consistency, but application developers should always consider the needs of their application before setting
read preference.
For more information on read preference or on the read preference modes, see Read Preference (page 406) and Read
Preference Modes (page 476).

2.2.2 Write Operations

Write Operations Overview (page 43) Provides an overview of MongoDBs data insertion and modification operations, including aspects of the syntax, and behavior.
42

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Write Concern (page 47) Describes the kind of guarantee MongoDB provides when reporting on the success of a
write operation.
Distributed Write Operations (page 51) Describes how MongoDB directs write operations on sharded clusters and
replica sets and the performance characteristics of these operations.
Write Operation Performance (page 52) Introduces the performance constraints and factors for writing data to MongoDB deployments.
Bulk Inserts in MongoDB (page 56) Describe behaviors associated with inserting an array of documents.
Record Padding (page 57) When storing documents on disk, MongoDB reserves space to allow documents to grow
efficiently during subsequent updates.
Write Operations Overview
A write operation is any operation that creates or modifies data in the MongoDB instance. In MongoDB, write
operations target a single collection. All write operations in MongoDB are atomic on the level of a single document.
There are three classes of write operations in MongoDB: insert, update, and remove. Insert operations add new data to
a collection. Update operations modify existing data, and remove operations delete data from a collection. No insert,
update, or remove can affect more than one document atomically.
For the update and remove operations, you can specify criteria, or conditions, that identify the documents to update or
remove. These operations use the same query syntax to specify the criteria as read operations (page 31).
After issuing these modification operations, MongoDB allows applications to determine the level of acknowledgment
returned from the database. See Write Concern (page 47).
Create

Create operations add new documents to a collection. In MongoDB, the db.collection.insert() method
performs create operations.
The following diagram highlights the components of a MongoDB insert operation:

Figure 2.13: The components of a MongoDB insert operations.

The following diagram shows the same query in SQL:
Example

2.2. MongoDB CRUD Concepts

43

MongoDB Documentation, Release 2.4.9

Figure 2.14: The components of a SQL INSERT statement.

The following operation inserts a new documents into the users collection. The new document has four fields name,
age, and status, and an _id field. MongoDB always adds the _id field to the new document if that field does not
exist.
db.users.insert(
{
name: "sue",
age: 26,
status: "A"
}
)

This operation inserts a new document into the users collection. The new document has four fields: name, age,
status, and an _id field. MongoDB always adds the _id field to a new document if the field does not exist.
For more information, see db.collection.insert() and Insert Documents (page 59).
Some updates also create records. If an update operation specifies the upsert flag and there are no documents that
match the query portion of the update operation, then MongoDB will convert the update into an insert.
With an upsert, applications can decide between performing an update or an insert operation using just a single call.
Both the update() method and the save() method can perform an upsert. See update() and save() for
details on performing an upsert with these methods.
See
SQL to MongoDB Mapping Chart (page 85) for additional examples of MongoDB write operations and the corresponding SQL statements.

Insert Behavior If you add a new document without the _id field, the client library or the mongod instance adds an
_id field and populates the field with a unique ObjectId.
If you specify the _id field, the value must be unique within the collection. For operations with write concern
(page 47), if you try to create a document with a duplicate _id value, mongod returns a duplicate key exception.
Update

Update operations modify existing documents in a collection. In MongoDB, db.collection.update() and

the db.collection.save() methods perform update operations. The db.collection.update() method
can accept query criteria to determine which documents to update as well as an option to update multiple rows. The
method can also accept options that affect its behavior such as the multi option to update multiple documents.
The following diagram highlights the components of a MongoDB update operation:

44

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Figure 2.15: The components of a MongoDB update operation.

The following diagram shows the same query in SQL:

Figure 2.16: The components of a SQL UPDATE statement.

Example
db.users.update(
{ age: { $gt: 18 } },
{ $set: { status: "A" } },
{ multi: true }
)

This update operation on the users collection sets the status field to A for the documents that match the criteria
of age greater than 18.
For more information, see db.collection.update() and db.collection.save(), and Modify Documents (page 67) for examples.
Update Behavior By default, the db.collection.update() method updates a single document. However,
with the multi option, update() can update all documents in a collection that match a query.
The db.collection.update() method either updates specific fields in the existing document or replaces the
document. See db.collection.update() for details.
When performing update operations that increase the document size beyond the allocated space for that document, the
update operation relocates the document on disk and may reorder the document fields depending on the type of update.
The db.collection.save() method replaces a document and can only update a single document.
db.collection.save() and Insert Documents (page 59) for more information

2.2. MongoDB CRUD Concepts

See

45

MongoDB Documentation, Release 2.4.9

Delete

Delete operations remove documents from a collection. In MongoDB, db.collection.remove() method performs delete operations. The db.collection.remove() method can accept query criteria to determine which
documents to remove.
The following diagram highlights the components of a MongoDB remove operation:

Figure 2.17: The components of a MongoDB remove operation.

The following diagram shows the same query in SQL:

Figure 2.18: The components of a SQL DELETE statement.

Example
db.users.remove(
{ status: "D" }
)

This delete operation on the users collection removes all documents that match the criteria of status equal to D.
For more information, see db.collection.remove() method and Remove Documents (page 68).
Remove Behavior By default, db.collection.remove() method removes all documents that match its query.
However, the method can accept a flag to limit the delete operation to a single document.
Isolation of Write Operations

The modification of a single document is always atomic, even if the write operation modifies multiple sub-documents
within that document. For write operations that modify multiple documents, the operation as a whole is not atomic,
and other operations may interleave.
No other operations are atomic. You can, however, attempt to isolate a write operation that affects multiple documents
using the isolation operator.
To isolate a sequence of write operations from other read and write operations, see Perform Two Phase Commits
(page 69).

46

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Write Concern
Write concern describes the guarantee that MongoDB provides when reporting on the success of a write operation.
The strength of the write concerns determine the level of guarantee. When inserts, updates and deletes have a weak
write concern, write operations return quickly. In some failure cases, write operations issued with weak write concerns
may not persist. With stronger write concerns, clients wait after sending a write operation for MongoDB to confirm
the write operations.
MongoDB provides different levels of write concern to better address the specific needs of applications. Clients
may adjust write concern to ensure that the most important operations persist successfully to an entire MongoDB
deployment. For other less critical operations, clients can adjust the write concern to ensure faster performance rather
than ensure persistence to the entire deployment.
See also:
Write Concern Reference (page 83) for a reference of specific write concern configuration. Also consider Write
Operations (page 42) for a general overview of write operations with MongoDB and Write Concern for Replica Sets
(page 402) for considerations specific to replica sets.
Note: The driver write concern (page 634) change created a new connection class in all of the MongoDB drivers.
The new class, called MongoClient, changed the default write concern. See the release notes (page 634) for this
change and the release notes for your driver.

Write Concern Levels

Clients issue write operations with some level of write concern. MongoDB has the following levels of conceptual
write concern, listed from weakest to strongest:
Errors Ignored With an errors ignored write concern, MongoDB does not acknowledge write operations. With
this level of write concern, the client cannot detect failed write operations. These errors include connection errors and
mongod exceptions such as duplicate key exceptions for unique indexes (page 334). Although the errors ignored write
concern provides fast performance, this performance gain comes at the cost of significant risks for data persistence
and durability.
To set errors ignored write concern, specify w values of -1 to your driver.
Warning: Do not use errors ignored write concern in normal operation.

Unacknowledged With an unacknowledged write concern, MongoDB does not acknowledge the receipt of write
operation. Unacknowledged is similar to errors ignored; however, drivers attempt to receive and handle network
errors when possible. The drivers ability to detect network errors depends on the systems networking configuration.
To set unacknowledged write concern, specify w values of 0 to your driver.
Before the releases outlined in Default Write Concern Change (page 634), this was the default write concern.
Acknowledged With a receipt acknowledged write concern, the mongod confirms the receipt of the write operation.
Acknowledged write concern allows clients to catch network, duplicate key, and other errors.
To set acknowledged write concern, specify w values of 1 to your driver.
MongoDB uses acknowledged write concern by default, after the releases outlined in Default Write Concern Change
(page 634).

2.2. MongoDB CRUD Concepts

47

MongoDB Documentation, Release 2.4.9

Figure 2.19: Write operation to a mongod instance with write concern of unacknowledged. The client does not
wait for any acknowledgment.

Figure 2.20: Write operation to a mongod instance with write concern of acknowledged. The client waits for
acknowledgment of success or exception.

48

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Internally, the default write concern calls getLastError with no arguments. For replica sets, you can define the
default write concern settings in the getLastErrorDefaults (page 470). When getLastErrorDefaults
(page 470) does not define a default write concern setting, getLastError defaults to basic receipt acknowledgment.
Journaled With a journaled write concern, the mongod acknowledges the write operation only after committing
the data to the journal. This write concern ensures that MongoDB can recover the data following a shutdown or power
interruption.
To set a journaled write concern, specify w values of 1 and set the journal or j option to true for your driver. You
must have journaling enabled to use this write concern.
With a journaled write concern, write operations must wait for the next journal commit. To reduce latency
for these operations, you can increase the frequency that MongoDB commits operations to the journal. See
journalCommitInterval for more information.

Figure 2.21: Write operation to a mongod instance with write concern of journaled. The mongod sends acknowledgment after it commits the write operation to the journal.

Note: Requiring journaled write concern in a replica set only requires a journal commit of the write operation to the
primary of the set regardless of the level of replica acknowledged write concern.

Replica Acknowledged Replica sets add several considerations for write concern. Basic write concerns affect write
operations on only one mongod instance. The w argument to getLastError provides replica acknowledged write
concerns. With replica acknowledged you can guarantee that the write operation propagates to the members of a
replica set. See Write Concern Reference (page 83) document for the values for w and Write Concern for Replica Sets
(page 402) for more information.
To set replica acknowledged write concern, specify w values greater than 1 to your driver.
Note: Requiring journaled write concern in a replica set only requires a journal commit of the write operation to the
primary of the set regardless of the level of replica acknowledged write concern.
2.2. MongoDB CRUD Concepts

49

MongoDB Documentation, Release 2.4.9

Figure 2.22: Write operation to a replica set with write concern level of w:2 or write to the primary and at least one
secondary.

50

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Distributed Write Operations

Write Operations on Sharded Clusters

For sharded collections in a sharded cluster, the mongos directs write operations from applications to the shards that
are responsible for the specific portion of the data set. The mongos uses the cluster metadata from the config database
(page 488) to route the write operation to the appropriate shards.

Figure 2.23: Diagram of a sharded cluster.

MongoDB partitions data in a sharded collection into ranges based on the values of the shard key. Then, MongoDB
distributes these chunks to shards. The shard key determines the distribution of chunks to shards. This can affect the
performance of write operations in the cluster.
Important: Update operations that affect a single document must include the shard key or the _id field. Updates
that affect multiple documents are more efficient in some situations if they have the shard key, but can be broadcast to
all shards.

2.2. MongoDB CRUD Concepts

51

MongoDB Documentation, Release 2.4.9

Figure 2.24: Diagram of the shard key value space segmented into smaller ranges or chunks.
If the value of the shard key increases or decreases with every insert, all insert operations target a single shard. As a
result, the capacity of a single shard becomes the limit for the insert capacity of the sharded cluster.
For more information, see Sharded Cluster Tutorials (page 506) and Bulk Inserts in MongoDB (page 56).
Write Operations on Replica Sets

In replica sets, all write operations go to the sets primary, which applies the write operation then records the operations on the primarys operation log or oplog. The oplog is a reproducible sequence of operations to the data set.
Secondary members of the set are continuously replicating the oplog and applying the operations to themselves in an
asynchronous process.
Large volumes of write operations, particularly bulk operations, may create situations where the secondary members
have difficulty applying the replicating operations from the primary at a sufficient rate: this can cause the secondarys
state to fall behind that of the primary. Secondaries that are significantly behind the primary present problems for
normal operation of the replica set, particularly failover (page 397) in the form of rollbacks (page 401) as well as
general read consistency (page 402).
To help avoid this issue, you can customize the write concern (page 47) to return confirmation of the write operation
to another member 4 of the replica set every 100 or 1,000 operations. This provides an opportunity for secondaries
to catch up with the primary. Write concern can slow the overall progress of write operations but ensure that the
secondaries can maintain a largely current state with respect to the primary.
For more information on replica sets and write operations, see Replica Acknowledged (page 49), Oplog Size (page 411),
and Change the Size of the Oplog (page 445).
Write Operation Performance
Indexes

After every insert, update, or delete operation, MongoDB must update every index associated with the collection in
addition to the data itself. Therefore, every index on a collection adds some amount of overhead for the performance
of write operations. 5
4 Calling getLastError intermittently with a w value of 2 or majority will slow the throughput of write traffic; however, this practice will
allow the secondaries to remain current with the state of the primary.
5 For inserts and updates to un-indexed fields, the overhead for sparse indexes (page 335) is less than for non-sparse indexes. Also for non-sparse
indexes, updates that do not change the record size have less indexing overhead.

52

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Figure 2.25: Diagram of default routing of reads and writes to the primary.

2.2. MongoDB CRUD Concepts

53

MongoDB Documentation, Release 2.4.9

Figure 2.26: Write operation to a replica set with write concern level of w:2 or write to the primary and at least one
secondary.

54

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

In general, the performance gains that indexes provide for read operations are worth the insertion penalty. However,
in order to optimize write performance when possible, be careful when creating new indexes and evaluate the existing
indexes to ensure that your queries actually use these indexes.
For indexes and queries, see Query Optimization (page 36). For more information on indexes, see Indexes (page 313)
and Indexing Strategies (page 368).
Document Growth

If an update operation causes a document to exceed the currently allocated record size, MongoDB relocates the document on disk with enough contiguous space to hold the document. These relocations take longer than in-place updates,
particularly if the collection has indexes. If a collection has indexes, MongoDB must update all index entries. Thus,
for a collection with many indexes, the move will impact the write throughput.
Some update operations, such as the $inc operation, do not cause an increase in document size. For these update
operations, MongoDB can apply the updates in-place. Other update operations, such as the $push operation, change
the size of the document.
In-place-updates are significantly more efficient than updates that cause document growth. When possible, use data
models (page 99) that minimize the need for document growth.
See Record Padding (page 57) for more information.
Storage Performance

Hardware The capability of the storage system creates some important physical limits for the performance of MongoDBs write operations. Many unique factors related to the storage system of the drive affect write performance,
including random access patterns, disk caches, disk readahead and RAID configurations.
Solid state drives (SSDs) can outperform spinning hard disks (HDDs) by 100 times or more for random workloads.
See
Production Notes (page 153) for recommendations regarding additional hardware and configuration options.

Journaling MongoDB uses write ahead logging to an on-disk journal to guarantee write operation (page 42) durability and to provide crash resiliency. Before applying a change to the data files, MongoDB writes the change operation
to the journal.
While the durability assurance provided by the journal typically outweigh the performance costs of the additional write
operations, consider the following interactions between the journal and performance:
if the journal and the data file reside on the same block device, the data files and the journal may have to contend
for a finite number of available write operations. Moving the journal to a separate device may increase the
capacity for write operations.
if applications specify write concern (page 47) that includes journaled (page 49), mongod will decrease the
duration between journal commits, which can increases the overall write load.
the duration between journal commits is configurable using the journalCommitInterval run-time option.
Decreasing the period between journal commits will increase the number of write operations, which can limit
MongoDBs capacity for write operations. Increasing the amount of time between commits may decrease the
total number of write operation, but also increases the chance that the journal will not record a write operation
in the event of a failure.
For additional information on journaling, see Journaling Mechanics (page 234).

2.2. MongoDB CRUD Concepts

55

MongoDB Documentation, Release 2.4.9

Bulk Inserts in MongoDB

In some situations you may need to insert or ingest a large amount of data into a MongoDB database. These bulk
inserts have some special considerations that are different from other write operations.
Use the insert() Method

The insert() method, when passed an array of documents, performs a bulk insert, and inserts each document
atomically. Bulk inserts can significantly increase performance by amortizing write concern (page 47) costs.
New in version 2.2: insert() in the mongo shell gained support for bulk inserts in version 2.2.
In the drivers (page 95), you can configure write concern for batches rather than on a per-document level.
Drivers have a ContinueOnError option in their insert operation, so that the bulk operation will continue to insert
remaining documents in a batch even if an insert fails.
Note: If multiple errors occur during a bulk insert, clients only receive the last error generated.
See also:
Driver documentation (page 95) for details on performing bulk inserts in your application. Also see Import and Export
MongoDB Data (page 150).
Bulk Inserts on Sharded Clusters

While ContinueOnError is optional on unsharded clusters, all bulk operations to a sharded collection run with
ContinueOnError, which cannot be disabled.
Large bulk insert operations, including initial data inserts or routine data import, can affect sharded cluster performance. For bulk inserts, consider the following strategies:
Pre-Split the Collection If the sharded collection is empty, then the collection has only one initial chunk, which
resides on a single shard. MongoDB must then take time to receive data, create splits, and distribute the split chunks
to the available shards. To avoid this performance cost, you can pre-split the collection, as described in Split Chunks
in a Sharded Cluster (page 537).
Insert to Multiple mongos To parallelize import processes, send insert operations to more than one mongos
instance. Pre-split empty collections first as described in Split Chunks in a Sharded Cluster (page 537).
Avoid Monotonic Throttling If your shard key increases monotonically during an insert, then all inserted data goes
to the last chunk in the collection, which will always end up on a single shard. Therefore, the insert capacity of the
cluster will never exceed the insert capacity of that single shard.
If your insert volume is larger than what a single shard can process, and if you cannot avoid a monotonically increasing
shard key, then consider the following modifications to your application:
Reverse the binary bits of the shard key. This preserves the information and avoids correlating insertion order
with increasing sequence of values.
Swap the first and last 16-bit words to shuffle the inserts.

56

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Example
The following example, in C++, swaps the leading and trailing 16-bit word of BSON ObjectIds generated so that they
are no longer monotonically increasing.
using namespace mongo;
OID make_an_id() {
OID x = OID::gen();
const unsigned char *p = x.getData();
swap( (unsigned short&) p[0], (unsigned short&) p[10] );
return x;
}
void foo() {
// create an object
BSONObj o = BSON( "_id" << make_an_id() << "x" << 3 << "name" << "jane" );
// now we may insert o into a sharded collection
}

See also:
Shard Keys (page 492) for information on choosing a sharded key. Also see Shard Key Internals (page 492) (in
particular, Choosing a Shard Key (page 511)).
Record Padding
Update operations can increase the size of the document 6 . If a document outgrows its current allocated record space,
MongoDB must allocate a new space and move the document to this new location.
To reduce the number of moves, MongoDB includes a small amount of extra space, or padding, when allocating the
record space. This padding reduces the likelihood that a slight increase in document size will cause the document to
exceed its allocated record size.
See also:
Write Operation Performance (page 52).
Padding Factor

To minimize document movements and their impact, MongoDB employs padding. MongoDB adaptively adjusts the
size of record allocations in a collection by adding a paddingFactor so that the documents have room to grow.
The paddingFactor indicates the padding for new inserts and moves.
To check the current paddingFactor on a collection, you can run the db.collection.stats() operation in
the mongo shell, as in the following example:
db.myCollection.stats()

Since MongoDB writes each document at a different point in time, the padding for each document will not be the
same. You can calculate the padding size by subtracting 1 from the paddingFactor, for example:
padding size = (paddingFactor - 1) * <document size>.

For example, a paddingFactor of 1.0 specifies no padding whereas a paddingFactor of 1.5 specifies a padding
size of 0.5 or 50 percent (50%) of the document size.
6

Documents in MongoDB can grow up to the full maximum BSON document size.

2.2. MongoDB CRUD Concepts

57

MongoDB Documentation, Release 2.4.9

Because the paddingFactor is relative to the size of each document, you cannot calculate the exact amount of
padding for a collection based on the average document size and padding factor.
If an update operation causes the document to decrease in size, for instance if you perform an $unset or a $pop
update, the document remains in place and effectively has more padding. If the document remains this size, the space
is not reclaimed until you perform a compact or a repairDatabase operation.
Operations That Remove Padding

The following operations remove padding: compact, repairDatabase, and initial replica sync operations. However, with the compact command, you can run the command with a paddingFactor or a paddingBytes
parameter. See compact command for details.
Padding is also removed if you use mongoexport a collection. If you use mongoimport into a new collection, mongoimport will not add padding. If you use mongoimport with an existing collection with padding,
mongoimport will not affect the existing padding.
When a database operation removes padding from a collection, subsequent updates to the collection that increase the
record size will have reduced throughput until the collections padding factor grows. However, the collection will
require less storage.
Record Allocation Strategies

New in version 2.2: collMod and usePowerOf2Sizes.

To more efficiently reuse the space freed as a result of deletions or document relocations, you can specify that MongoDB allocates record sizes in powers of 2. To do so, use the collMod command with the usePowerOf2Sizes
flag. See collMod command for more details. As with all padding, power of 2 size allocations minimizes, but does
not eliminate, document movements.
See also Can I manually pad documents to prevent moves during updates? (page 567)

2.3 MongoDB CRUD Tutorials

The following tutorials provide instructions for querying and modifying data. For a higher-level overview of these
operations, see MongoDB CRUD Operations (page 27).
Insert Documents (page 59) Insert new documents into a collection.
Query Documents (page 60) Find documents in a collection using search criteria.
Limit Fields to Return from a Query (page 64) Limit which fields are returned by a query.
Iterate a Cursor in the mongo Shell (page 65) Access documents returned by a find query by iterating the cursor,
either manually or using the iterator index.
Analyze Query Performance (page 66) Analyze the efficiency of queries and determine how a query uses available
indexes.
Modify Documents (page 67) Modify documents in a collection
Remove Documents (page 68) Remove documents from a collection.
Perform Two Phase Commits (page 69) Use two-phase commits when writing data to multiple documents.
Create Tailable Cursor (page 74) Create tailable cursors for use in capped collections with high numbers of write
operations for which an index would be too expensive.

58

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Isolate Sequence of Operations (page 76) Use the <isolation> isolated operator to isolate a single write
operation that affects multiple documents, preventing other operations from interrupting the sequence of write
operations.
Create an Auto-Incrementing Sequence Field (page 78) Describes how to create an incrementing sequence number
for the _id field using a Counters Collection or an Optimistic Loop.
Limit Number of Elements in an Array after an Update (page 81) Use $push with various modifiers to sort and
maintain an array of fixed size after update

2.3.1 Insert Documents

In MongoDB, the db.collection.insert() method adds new documents into a collection. In addition, both the
db.collection.update() method and the db.collection.save() method can also add new documents
through an operation called an upsert. An upsert is an operation that performs either an update of an existing document
or an insert of a new document if the document to modify does not exist.
This tutorial provides examples of insert operations using each of the three methods in the mongo shell.
Insert a Document with insert() Method
The following statement inserts a document with three fields into the collection inventory:
db.inventory.insert( { _id: 10, type: "misc", item: "card", qty: 15 } )

In the example, the document has a user-specified _id field value of 10. The value must be unique within the
inventory collection.
For more examples, see insert().
Insert a Document with update() Method
Call the update() method with the upsert flag to create a new document if no document matches the updates
query criteria. 7
The following example creates a new document if no document in the inventory collection contains { type:
"books", item : "journal" }:
db.inventory.update(
{ type: "book", item : "journal" },
{ $set : { qty: 10 } },
{ upsert : true }
)

MongoDB adds the _id field and assigns as its value a unique ObjectId. The new document includes the item and
type fields from the <query> criteria and the qty field from the <update> parameter.
{ "_id" : ObjectId("51e8636953dbe31d5f34a38a"), "item" : "journal", "qty" : 10, "type" : "book" }

For more examples, see update().

7 Prior to version 2.2, in the mongo shell, you would specify the upsert and the multi options in the update() method as positional
boolean options. See update() for details.

2.3. MongoDB CRUD Tutorials

59

MongoDB Documentation, Release 2.4.9

Insert a Document with save() Method

To insert a document with the save() method, pass the method a document that does not contain the _id field or a
document that contains an _id field that does not exist in the collection.
The following example creates a new document in the inventory collection:
db.inventory.save( { type: "book", item: "notebook", qty: 40 } )

MongoDB adds the _id field and assigns as its value a unique ObjectId.
{ "_id" : ObjectId("51e866e48737f72b32ae4fbc"), "type" : "book", "item" : "notebook", "qty" : 40 }

For more examples, see save().

2.3.2 Query Documents

In MongoDB, the db.collection.find() method retrieves documents from a collection.
db.collection.find() method returns a cursor (page 35) to the retrieved documents.

The

This tutorial provides examples of read operations using the db.collection.find() method in the mongo
shell. In these examples, the retrieved documents contain all their fields. To restrict the fields to return in the retrieved
documents, see Limit Fields to Return from a Query (page 64).
Select All Documents in a Collection
An empty query document ({}) selects all documents in the collection:
db.inventory.find( {} )

Not specifying a query document to the find() is equivalent to specifying an empty query document. Therefore the
following operation is equivalent to the previous operation:
db.inventory.find()

Specify Equality Condition

To specify equality condition, use the query document { <field>:
contain the <field> with the specified <value>.

<value> } to select all documents that

The following example retrieves from the inventory collection all documents where the type field has the value
snacks:
db.inventory.find( { type: "snacks" } )

Specify Conditions Using Query Operators

A query document can use the query operators to specify conditions in a MongoDB query.
The following example selects all documents in the inventory collection where the value of the type field is either
food or snacks:
8 The db.collection.findOne() method also performs a read operation to return a single document.
db.collection.findOne() method is the db.collection.find() method with a limit of 1.

60

Internally, the

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

db.inventory.find( { type: { $in: [ 'food', 'snacks' ] } } )

Although you can express this query using the $or operator, use the $in operator rather than the $or operator when
performing equality checks on the same field.
Refer to the http://docs.mongodb.org/manualreference/operator document for the complete list
of query operators.
Specify AND Conditions
A compound query can specify conditions for more than one field in the collections documents. Implicitly, a logical
AND conjunction connects the clauses of a compound query so that the query selects the documents in the collection
that match all the conditions.
In the following example, the query document specifies an equality match on the field food and a less than ($lt)
comparison match on the field price:
db.inventory.find( { type: 'food', price: { $lt: 9.95 } } )

This query selects all documents where the type field has the value food and the value of the price field is less
than 9.95. See comparison operators for other comparison operators.
Specify OR Conditions
Using the $or operator, you can specify a compound query that joins each clause with a logical OR conjunction so
that the query selects the documents in the collection that match at least one condition.
In the following example, the query document selects all documents in the collection where the field qty has a value
greater than ($gt) 100 or the value of the price field is less than ($lt) 9.95:
db.inventory.find(
{ $or: [
{ qty: { $gt: 100 } },
{ price: { $lt: 9.95 } }
]
}
)

Specify AND as well as OR Conditions

With additional clauses, you can specify precise conditions for matching documents.
In the following example, the compound query document selects all documents in the collection where the value of
the type field is food and either the qty has a value greater than ($gt) 100 or the value of the price field is
less than ($lt) 9.95:
db.inventory.find( { type: 'food', $or: [ { qty: { $gt: 100 } },
{ price: { $lt: 9.95 } } ]
} )

Subdocuments
When the field holds an embedded document (i.e. subdocument), you can either specify the entire subdocument as
the value of a field, or reach into the subdocument using dot notation, to specify values for individual fields in the
subdocument:
2.3. MongoDB CRUD Tutorials

61

MongoDB Documentation, Release 2.4.9

Exact Match on Subdocument

To specify an equality match on the whole subdocument, use the query document { <field>: <value> }
where <value> is the subdocument to match. Equality matches on a subdocument require that the subdocument
field match exactly the specified <value>, including the field order.
In the following example, the query matches all documents where the value of the field producer is a subdocument that contains only the field company with the value ABC123 and the field address with the value 123
Street, in the exact order:
db.inventory.find(
{
producer: {
company: 'ABC123',
address: '123 Street'
}
}
)

Equality Match on Fields within Subdocument

Equality matches for specific fields within subdocuments select the documents in the collection when the field in the
subdocument contains a field that matches the specified value.
In the following example, the query uses the dot notation to match all documents where the value of the field
producer is a subdocument that contains a field company with the value ABC123 and may contain other fields:
db.inventory.find( { 'producer.company': 'ABC123' } )

Arrays
When the field holds an array, you can query for an exact array match or for specific values in the array. If the array
holds sub-documents, you can query for specific fields within the sub-documents using dot notation:
Exact Match on an Array

To specify equality match on an array, use the query document { <field>: <value> } where <value> is
the array to match. Equality matches on the array require that the array field match exactly the specified <value>,
including the element order.
In the following example, the query matches all documents where the value of the field tags is an array that holds
exactly three elements, fruit, food, and citrus, in this order:
db.inventory.find( { tags: [ 'fruit', 'food', 'citrus' ] } )

Match an Array Element

Equality matches can specify a single element in the array to match. These specifications match if the array contains
at least one element with the specified value.
In the following example, the query matches all documents where the value of the field tags is an array that contains
fruit as one of its elements:

62

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

db.inventory.find( { tags: 'fruit' } )

Match a Specific Element of an Array

Equality matches can specify equality matches for an element at a particular index or position of the array.
In the following example, the query uses the dot notation to match all documents where the value of the tags field is
an array whose first element equals fruit:
db.inventory.find( { 'tags.0' : 'fruit' } )

Array of Subdocuments

Match a Field in the Subdocument Using the Array Index If you know the array index of the subdocument, you
can specify the document using the subdocuments position.
The following example selects all documents where the memos contains an array whose first element (i.e. index is 0)
is a subdocument with the field by with the value shipping:
db.inventory.find( { 'memos.0.by': 'shipping' } )

Match a Field Without Specifying Array Index If you do not know the index position of the subdocument, concatenate the name of the field that contains the array, with a dot (.) and the name of the field in the subdocument.
The following example selects all documents where the memos field contains an array that contains at least one
subdocument with the field by with the value shipping:
db.inventory.find( { 'memos.by': 'shipping' } )

Match Multiple Fields To match by multiple fields in the subdocument, you can use either dot notation or the
$elemMatch operator:
The following example uses dot notation to query for documents where the value of the memos field is an array that has
at least one subdocument that contains the field memo equal to on time and the field by equal to shipping:
db.inventory.find(
{
'memos.memo': 'on time',
'memos.by': 'shipping'
}
)

The following example uses $elemMatch to query for documents where the value of the memos field is an array that has at least one subdocument that contains the field memo equal to on time and the field by equal to
shipping:
db.inventory.find( {
memos: {
$elemMatch: {
memo : 'on time',
by: 'shipping'
}
}
}
)

2.3. MongoDB CRUD Tutorials

63

MongoDB Documentation, Release 2.4.9

2.3.3 Limit Fields to Return from a Query

The projection specification limits the fields to return for all matching documents. The projection takes the form of a
document with a list of fields for inclusion or exclusion from the result set. You can either specify the fields to include
(e.g. { field: 1 }) or specify the fields to exclude (e.g. { field: 0 }).
Important: The _id field is, by default, included in the result set. To exclude the _id field from the result set, you
need to specify in the projection document the exclusion of the _id field (i.e. { _id: 0 }).
You cannot combine inclusion and exclusion semantics in a single projection with the exception of the _id field.
This tutorial offers various query examples that limit the fields to return for all matching documents. The examples in
this tutorial use a collection inventory and use the db.collection.find() method in the mongo shell. The
db.collection.find() method returns a cursor (page 35) to the retrieved documents. For examples on query
selection criteria, see Query Documents (page 60).
Return All Fields in Matching Documents
If you specify no projection, the find() method returns all fields of all documents that match the query.
db.inventory.find( { type: 'food' } )

This operation will return all documents in the inventory collection where the value of the type field is food.
The returned documents contain all its fields.
Return the Specified Fields and the _id Field Only
A projection can explicitly include several fields. In the following operation, find() method returns all documents
that match the query. In the result set, only the item and qty fields and, by default, the _id field return in the
matching documents.
db.inventory.find( { type: 'food' }, { item: 1, qty: 1 } )

Return Specified Fields Only

You can remove the _id field from the results by specifying its exclusion in the projection, as in the following
example:
db.inventory.find( { type: 'food' }, { item: 1, qty: 1, _id:0 } )

This operation returns all documents that match the query. In the result set, only the item and qty fields return in
the matching documents.
Return All But the Excluded Field
To exclude a single field or group of fields you can use a projection in the following form:
db.inventory.find( { type: 'food' }, { type:0 } )

This operation returns all documents where the value of the type field is food. In the result set, the type field does
not return in the matching documents.
With the exception of the _id field you cannot combine inclusion and exclusion statements in projection documents.

64

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Projection for Array Fields

The $elemMatch and $slice projection operators are the only way to project portions of an array.
Tip
MongoDB does not support projections of portions of arrays except when using the $elemMatch and $slice
projection operators.

2.3.4 Iterate a Cursor in the mongo Shell

The db.collection.find() method returns a cursor. To access the documents, you need to iterate the cursor.
However, in the mongo shell, if the returned cursor is not assigned to a variable using the var keyword, then the
cursor is automatically iterated up to 20 times to print up to the first 20 documents in the results. The following
describes ways to manually iterate the cursor to access the documents or to use the iterator index.
Manually Iterate the Cursor
In the mongo shell, when you assign the cursor returned from the find() method to a variable using the var
keyword, the cursor does not automatically iterate.
You can call the cursor variable in the shell to iterate up to 20 times
following example:

and print the matching documents, as in the

var myCursor = db.inventory.find( { type: 'food' } );

myCursor

You can also use the cursor method next() to access the documents, as in the following example:
var myCursor = db.inventory.find( { type: 'food' } );
var myDocument = myCursor.hasNext() ? myCursor.next() : null;
if (myDocument) {
var myItem = myDocument.item;
print(tojson(myItem));
}

As an alternative print operation, consider the printjson() helper method to replace print(tojson()):
if (myDocument) {
var myItem = myDocument.item;
printjson(myItem);
}

You can use the cursor method forEach() to iterate the cursor and access the documents, as in the following
example:
var myCursor =

db.inventory.find( { type: 'food' } );

myCursor.forEach(printjson);

See JavaScript cursor methods and your driver (page 95) documentation for more information on cursor methods.
9

You can use the DBQuery.shellBatchSize to change the number of iteration from the default value 20. See Executing Queries
(page 213) for more information.

2.3. MongoDB CRUD Tutorials

65

MongoDB Documentation, Release 2.4.9

Iterator Index
In the mongo shell, you can use the toArray() method to iterate the cursor and return the documents in an array,
as in the following:
var myCursor = db.inventory.find( { type: 'food' } );
var documentArray = myCursor.toArray();
var myDocument = documentArray[3];

The toArray() method loads into RAM all documents returned by the cursor; the toArray() method exhausts
the cursor.
Additionally, some drivers (page 95) provide access to the documents by using an index on the cursor (i.e.
cursor[index]). This is a shortcut for first calling the toArray() method and then using an index on the
resulting array.
Consider the following example:
var myCursor = db.inventory.find( { type: 'food' } );
var myDocument = myCursor[3];

The myCursor[3] is equivalent to the following example:

myCursor.toArray() [3];

2.3.5 Analyze Query Performance

The explain() cursor method allows you to inspect the operation of the query system. This method is useful for
analyzing the efficiency of queries, and for determining how the query uses the index. The explain() method tests
the query operation, and not the timing of query performance. Because explain() attempts multiple query plans,
it does not reflect an accurate timing of query performance.
Evaluate the Performance of a Query
To use the explain() method, call the method on a cursor returned by find().
Example
Evaluate a query on the type field on the collection inventory that has an index on the type field.
db.inventory.find( { type: 'food' } ).explain()

Consider the results:

{
"cursor" : "BtreeCursor type_1",
"isMultiKey" : false,
"n" : 5,
"nscannedObjects" : 5,
"nscanned" : 5,
"nscannedObjectsAllPlans" : 5,
"nscannedAllPlans" : 5,
"scanAndOrder" : false,
"indexOnly" : false,
"nYields" : 0,
"nChunkSkips" : 0,
"millis" : 0,

66

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

"indexBounds" : { "type" : [
[ "food",
"food" ]
] },
"server" : "mongodbo0.example.net:27017" }

The BtreeCursor value of the cursor field indicates that the query used an index.
This query returned 5 documents, as indicated by the n field.
To return these 5 documents, the query scanned 5 documents from the index, as indicated by the nscanned field,
and then read 5 full documents from the collection, as indicated by the nscannedObjects field.
Without the index, the query would have scanned the whole collection to return the 5 documents.
See explain-results method for full details on the output.

Compare Performance of Indexes

To manually compare the performance of a query using more than one index, you can use the hint() and
explain() methods in conjunction.
Example
Evaluate a query using different indexes:
db.inventory.find( { type: 'food' } ).hint( { type: 1 } ).explain()
db.inventory.find( { type: 'food' } ).hint( { type: 1, name: 1 } ).explain()

These return the statistics regarding the execution of the query using the respective index.
Note: If you run explain() without including hint(), the query optimizer reevaluates the query and runs against
multiple indexes before returning the query statistics.
For more detail on the explain output, see explain-results.

2.3.6 Modify Documents

In MongoDB, both db.collection.update() and db.collection.save() modify existing documents in
a collection. db.collection.update() provides additional control over the modification. For example, you
can modify existing data or modify a group of documents that match a query with db.collection.update().
Alternately, db.collection.save() replaces an existing document with the same _id field.
This document provides examples of the update operations using each of the two methods in the mongo shell.
Modify Multiple Documents with update() Method
By default, the update() method updates a single document that matches its selection criteria. Call the method with
the multi option set to true to update multiple documents. 10
The following example finds all documents with type equal to "book" and modifies their qty field by -1. The
example uses $inc, which is one of the update operators available.
10

This shows the syntax for MongoDB 2.2 and later. For syntax for versions prior to 2.2, see update().

2.3. MongoDB CRUD Tutorials

67

MongoDB Documentation, Release 2.4.9

db.inventory.update(
{ type : "book" },
{ $inc : { qty : -1 } },
{ multi: true }
)

For more examples, see update().

Modify a Document with save() Method
The save() method can replace an existing document. To replace a document with the save() method, pass the
method a document with an _id field that matches an existing document.
The following example completely replaces the document with the _id equal to 10 in the inventory collection:
db.inventory.save(
{
_id: 10,
type: "misc",
item: "placard"
}
)

For further examples, see save().

2.3.7 Remove Documents

In MongoDB, the db.collection.remove() method removes documents from a collection. You can remove
all documents from a collection, remove all documents that match a condition, or limit the operation to remove just a
single document.
This tutorial provides examples of remove operations using the db.collection.remove() method in the mongo
shell.
Remove All Documents
If you do not specify a query, remove() removes all documents from a collection, but does not remove the indexes.
[#drop]_
The following example removes all documents from the inventory collection:
db.inventory.remove()

To remove all documents from a collection, it may be more efficient to use the drop() method to drop the entire
collection, including the indexes, and then recreate the collection and rebuild the indexes.
Remove Documents that Match a Condition
To remove the documents that match a deletion criteria, call the remove() method with the <query> parameter.
The following example removes all documents from the inventory collection where the type field equals food:
db.inventory.remove( { type : "food" } )

For large deletion operations, it may be more efficient to copy the documents that you want to keep to a new collection
and then use drop() on the original collection.
68

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Remove a Single Document that Matches a Condition

To remove a single document, call the remove() method with the justOne parameter set to true or 1.
The following example removes one document from the inventory collection where the type field equals food:
db.inventory.remove( { type : "food" }, 1 )

To delete a single document sorted by some specified order, use the findAndModify() method.

2.3.8 Perform Two Phase Commits

Synopsis
This document provides a pattern for doing multi-document updates or transactions using a two-phase commit
approach for writing data to multiple documents. Additionally, you can extend this process to provide a rollback
(page 72) like functionality.
Background
Operations on a single document are always atomic with MongoDB databases; however, operations that involve multiple documents, which are often referred to as transactions, are not atomic. Since documents can be fairly complex
and contain multiple nested documents, single-document atomicity provides necessary support for many practical
use cases.
Thus, without precautions, success or failure of the database operation cannot be all or nothing, and without support
for multi-document transactions its possible for an operation to succeed for some operations and fail with others.
When executing a transaction composed of several sequential operations the following issues arise:
Atomicity: if one operation fails, the previous operation within the transaction must rollback to the previous
state (i.e. the nothing, in all or nothing.)
Isolation: operations that run concurrently with the transaction operation set must see a consistent view of the
data throughout the transaction process.
Consistency: if a major failure (i.e. network, hardware) interrupts the transaction, the database must be able to
recover a consistent state.
Despite the power of single-document atomic operations, there are cases that require multi-document transactions. For
these situations, you can use a two-phase commit, to provide support for these kinds of multi-document updates.
Because documents can represent both pending data and states, you can use a two-phase commit to ensure that data is
consistent, and that in the case of an error, the state that preceded the transaction is recoverable (page 72).
Note: Because only single-document operations are atomic with MongoDB, two-phase commits can only offer
transaction-like semantics. Its possible for applications to return intermediate data at intermediate points during the
two-phase commit or rollback.

Pattern
Overview

The most common example of transaction is to transfer funds from account A to B in a reliable way, and this pattern
uses this operation as an example. In a relational database system, this operation would encapsulate subtracting funds

2.3. MongoDB CRUD Tutorials

69

MongoDB Documentation, Release 2.4.9

from the source (A) account and adding them to the destination (B) within a single atomic transaction. For MongoDB,
you can use a two-phase commit in these situations to achieve a compatible response.
All of the examples in this document use the mongo shell to interact with the database, and assume that you have two
collections: First, a collection named accounts that will store data about accounts with one account per document,
and a collection named transactions which will store the transactions themselves.
Begin by creating two accounts named A and B, with the following command:
db.accounts.save({name: "A", balance: 1000, pendingTransactions: []})
db.accounts.save({name: "B", balance: 1000, pendingTransactions: []})

To verify that these operations succeeded, use find():

db.accounts.find()

mongo will return two documents that resemble the following:

{ "_id" : ObjectId("4d7bc66cb8a04f512696151f"), "name" : "A", "balance" : 1000, "pendingTransactions"

{ "_id" : ObjectId("4d7bc67bb8a04f5126961520"), "name" : "B", "balance" : 1000, "pendingTransactions"

Transaction Description

Set Transaction State to Initial Create the transaction collection by inserting the following document. The
transaction document holds the source and destination, which refer to the name fields of the accounts
collection, as well as the value field that represents the amount of data change to the balance field. Finally, the
state field reflects the current state of the transaction.
db.transactions.save({source: "A", destination: "B", value: 100, state: "initial"})

To verify that these operations succeeded, use find():

db.transactions.find()

This will return a document similar to the following:

{ "_id" : ObjectId("4d7bc7a8b8a04f5126961522"), "source" : "A", "destination" : "B", "value" : 100, "

Switch Transaction State to Pending Before modifying either records in the accounts collection, set the transaction state to pending from initial.
Set the local variable t in your shell session, to the transaction document using findOne():
t = db.transactions.findOne({state: "initial"})

After assigning this variable t, the shell will return the value of t, you will see the following output:
{
"_id" : ObjectId("4d7bc7a8b8a04f5126961522"),
"source" : "A",
"destination" : "B",
"value" : 100,
"state" : "initial"
}

Use update() to change the value of state to pending:

db.transactions.update({_id: t._id}, {$set: {state: "pending"}})
db.transactions.find()

70

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

The find() operation will return the contents of the transactions collection, which should resemble the following:

{ "_id" : ObjectId("4d7bc7a8b8a04f5126961522"), "source" : "A", "destination" : "B", "value" : 100, "

Apply Transaction to Both Accounts Continue by applying the transaction to both accounts. The update()
query will prevent you from applying the transaction if the transaction is not already marked as pending. Use the
following update() operation:

db.accounts.update({name: t.source, pendingTransactions: {$ne: t._id}}, {$inc: {balance: -t.value}, $

db.accounts.update({name: t.destination, pendingTransactions: {$ne: t._id}}, {$inc: {balance: t.value
db.accounts.find()

The find() operation will return the contents of the accounts collection, which should now resemble the following:

{ "_id" : ObjectId("4d7bc97fb8a04f5126961523"), "balance" : 900, "name" : "A", "pendingTransactions"

{ "_id" : ObjectId("4d7bc984b8a04f5126961524"), "balance" : 1100, "name" : "B", "pendingTransactions"

Set Transaction State to Committed

committed:

Use the following update() operation to set the transactions state to

db.transactions.update({_id: t._id}, {$set: {state: "committed"}})

db.transactions.find()

The find() operation will return the contents of the transactions collection, which should now resemble the
following:

{ "_id" : ObjectId("4d7bc7a8b8a04f5126961522"), "destination" : "B", "source" : "A", "state" : "commi

Remove Pending Transaction Use the following update() operation to set remove the pending transaction from
the documents in the accounts collection:
db.accounts.update({name: t.source}, {$pull: {pendingTransactions: t._id}})
db.accounts.update({name: t.destination}, {$pull: {pendingTransactions: t._id}})
db.accounts.find()

The find() operation will return the contents of the accounts collection, which should now resemble the following:

{ "_id" : ObjectId("4d7bc97fb8a04f5126961523"), "balance" : 900, "name" : "A", "pendingTransactions"

{ "_id" : ObjectId("4d7bc984b8a04f5126961524"), "balance" : 1100, "name" : "B", "pendingTransactions"

Set Transaction State to Done Complete the transaction by setting the state of the transaction document to done:
db.transactions.update({_id: t._id}, {$set: {state: "done"}})
db.transactions.find()

The find() operation will return the contents of the transactions collection, which should now resemble the
following:

{ "_id" : ObjectId("4d7bc7a8b8a04f5126961522"), "destination" : "B", "source" : "A", "state" : "done"

2.3. MongoDB CRUD Tutorials

71

MongoDB Documentation, Release 2.4.9

Recovering from Failure Scenarios

The most important part of the transaction procedure is not, the prototypical example above, but rather the possibility
for recovering from the various failure scenarios when transactions do not complete as intended. This section will
provide an overview of possible failures and provide methods to recover from these kinds of events.
There are two classes of failures:
all failures that occur after the first step (i.e. setting the transaction set to initial (page 70)) but before the third
step (i.e. applying the transaction to both accounts (page 71).)
To recover, applications should get a list of transactions in the pending state and resume from the second step
(i.e. switching the transaction state to pending (page 70).)
all failures that occur after the third step (i.e. applying the transaction to both accounts (page 71)) but before
the fifth step (i.e. setting the transaction state to done (page 71).)
To recover, application should get a list of transactions in the committed state and resume from the fourth
step (i.e. remove the pending transaction (page 71).)
Thus, the application will always be able to resume the transaction and eventually arrive at a consistent state. Run
the following recovery operations every time the application starts to catch any unfinished transactions. You may also
wish run the recovery operation at regular intervals to ensure that your data remains in a consistent state.
The time required to reach a consistent state depends, on how long the application needs to recover each transaction.
Rollback In some cases you may need to rollback or undo a transaction when the application needs to cancel
the transaction, or because it can never recover as in cases where one of the accounts doesnt exist, or stops existing
during the transaction.
There are two possible rollback operations:
1. After you apply the transaction (page 71) (i.e. the third step), you have fully committed the transaction and you
should not roll back the transaction. Instead, create a new transaction and switch the values in the source and
destination fields.
2. After you create the transaction (page 70) (i.e. the first step), but before you apply the transaction (page 71) (i.e
the third step), use the following process:
Set Transaction State to Canceling
update() operation:

Begin by setting the transactions state to canceling using the following

db.transactions.update({_id: t._id}, {$set: {state: "canceling"}})

Undo the Transaction

counts:

Use the following sequence of operations to undo the transaction operation from both ac-

db.accounts.update({name: t.source, pendingTransactions: t._id}, {$inc: {balance: t.value}, $pull: {p

db.accounts.update({name: t.destination, pendingTransactions: t._id}, {$inc: {balance: -t.value}, $pu
db.accounts.find()

The find() operation will return the contents of the accounts collection, which should resemble the following:

{ "_id" : ObjectId("4d7bc97fb8a04f5126961523"), "balance" : 1000, "name" : "A", "pendingTransactions"

{ "_id" : ObjectId("4d7bc984b8a04f5126961524"), "balance" : 1000, "name" : "B", "pendingTransactions"

72

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Set Transaction State to Canceled Finally, use the following update() operation to set the transactions state to
canceled:
db.transactions.update({_id: t._id}, {$set: {state: "canceled"}})

Multiple Applications Transactions exist, in part, so that several applications can create and run operations concurrently without causing data inconsistency or conflicts. As a result, it is crucial that only one 1 application can handle
a given transaction at any point in time.
Consider the following example, with a single transaction (i.e. T1) and two applications (i.e. A1 and A2). If both
applications begin processing the transaction which is still in the initial state (i.e. step 1 (page 70)), then:
A1 can apply the entire whole transaction before A2 starts.
A2 will then apply T1 for the second time, because the transaction does not appear as pending in the accounts
documents.
To handle multiple applications, create a marker in the transaction document itself to identify the application that is
handling the transaction. Use findAndModify() method to modify the transaction:
t = db.transactions.findAndModify({query: {state: "initial", application: {$exists: 0}},
update: {$set: {state: "pending", application: "A1"}},
new: true})

When you modify and reassign the local shell variable t, the mongo shell will return the t object, which should
resemble the following:
{
"_id" : ObjectId("4d7be8af2c10315c0847fc85"),
"application" : "A1",
"destination" : "B",
"source" : "A",
"state" : "pending",
"value" : 150
}

Amend the transaction operations to ensure that only applications that match the identifier in the value of the
application field before applying the transaction.
If the application A1 fails during transaction execution, you can use the recovery procedures (page 72), but applications
should ensure that they owns the transaction before applying the transaction. For example to resume pending jobs,
use a query that resembles the following:
db.transactions.find({application: "A1", state: "pending"})

This will (or may) return a document from the transactions document that resembles the following:

{ "_id" : ObjectId("4d7be8af2c10315c0847fc85"), "application" : "A1", "destination" : "B", "source" :

Using Two-Phase Commits in Production Applications

The example transaction above is intentionally simple. For example, it assumes that:
it is always possible roll back operations on an account.
account balances can hold negative values.
Production implementations would likely be more complex. Typically accounts need information about current balance, pending credits, pending debits. Then:

2.3. MongoDB CRUD Tutorials

73

MongoDB Documentation, Release 2.4.9

when your application switches the transaction state to pending (page 70) (i.e. step 2) it would also make sure
that the account has sufficient funds for the transaction. During this update operation, the application would also
modify the values of the credits and debits as well as adding the transaction as pending.
when your application removes the pending transaction (page 71) (i.e. step 4) the application would apply the
transaction on balance, modify the credits and debits as well as removing the transaction from the pending
field., all in one update.
Because all of the changes in the above two operations occur within a single update() operation, these changes are
all atomic.
Additionally, for most important transactions, ensure that:
the database interface (i.e. client library or driver) has a reasonable write concern configured to ensure that
operations return a response on the success or failure of a write operation.
your mongod instance has journaling enabled to ensure that your data is always in a recoverable state, in the
event of an unclean mongod shutdown.

2.3.9 Create Tailable Cursor

Overview
By default, MongoDB will automatically close a cursor when the client has exhausted all results in the cursor. However, for capped collections (page 161) you may use a Tailable Cursor that remains open after the client exhausts
the results in the initial cursor. Tailable cursors are conceptually equivalent to the tail Unix command with the -f
option (i.e. with follow mode.) After clients insert new additional documents into a capped collection, the tailable
cursor will continue to retrieve documents.
Use tailable cursors on capped collections that have high write volumes where indexes arent practical. For instance,
MongoDB replication (page 377) uses tailable cursors to tail the primarys oplog.
Note: If your query is on an indexed field, do not use tailable cursors, but instead, use a regular cursor. Keep track of
the last value of the indexed field returned by the query. To retrieve the newly added documents, query the collection
again using the last value of the indexed field in the query criteria, as in the following example:
db.<collection>.find( { indexedField: { $gt: <lastvalue> } } )

Consider the following behaviors related to tailable cursors:

Tailable cursors do not use indexes and return documents in natural order.
Because tailable cursors do not use indexes, the initial scan for the query may be expensive; but, after initially
exhausting the cursor, subsequent retrievals of the newly added documents are inexpensive.
Tailable cursors may become dead, or invalid, if either:
the query returns no match.
the cursor returns the document at the end of the collection and then the application deletes those document.
A dead cursor has an id of 0.
See your driver documentation (page 95) for the driver-specific method to specify the tailable cursor. For more
information on the details of specifying a tailable cursor, see MongoDB wire protocol11 documentation.
11 http://docs.mongodb.org/meta-driver/latest/legacy/mongodb-wire-protocol

74

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

C++ Example
The tail function uses a tailable cursor to output the results from a query to a capped collection:
The function handles the case of the dead cursor by having the query be inside a loop.
To periodically check for new data, the cursor->more() statement is also inside a loop.
#include "client/dbclient.h"
using namespace mongo;
/*
* Example of a tailable cursor.
* The function "tails" the capped collection (ns) and output elements as they are added.
* The function also handles the possibility of a dead cursor by tracking the field 'insertDate'.
* New documents are added with increasing values of 'insertDate'.
*/
void tail(DBClientBase& conn, const char *ns) {
BSONElement lastValue = minKey.firstElement();
Query query = Query().hint( BSON( "$natural" << 1 ) );
while ( 1 ) {
auto_ptr<DBClientCursor> c =
conn.query(ns, query, 0, 0, 0,
QueryOption_CursorTailable | QueryOption_AwaitData );
while ( 1 ) {
if ( !c->more() ) {
if ( c->isDead() ) {
break;
}
continue;
}
BSONObj o = c->next();
lastValue = o["insertDate"];
cout << o.toString() << endl;
}
query = QUERY( "insertDate" << GT << lastValue ).hint( BSON( "$natural" << 1 ) );
}
}

The tail function performs the following actions:

Initialize the lastValue variable, which tracks the last accessed value. The function will use the lastValue
if the cursor becomes invalid and tail needs to restart the query. Use hint() to ensure that the query uses
the $natural order.
In an outer while(1) loop,
Query the capped collection and return a tailable cursor that blocks for several seconds waiting for new
documents

2.3. MongoDB CRUD Tutorials

75

MongoDB Documentation, Release 2.4.9

auto_ptr<DBClientCursor> c =
conn.query(ns, query, 0, 0, 0,
QueryOption_CursorTailable | QueryOption_AwaitData );

* Specify the capped collection using ns as an argument to the function.

* Set the QueryOption_CursorTailable option to create a tailable cursor.
* Set the QueryOption_AwaitData option so that the returned cursor blocks for a few seconds to
wait for data.
In an inner while (1) loop, read the documents from the cursor:
* If the cursor has no more documents and is not invalid, loop the inner while loop to recheck for
more documents.
* If the cursor has no more documents and is dead, break the inner while loop.
* If the cursor has documents:
output the document,
update the lastValue value,
and loop the inner while (1) loop to recheck for more documents.
If the logic breaks out of the inner while (1) loop and the cursor is invalid:
* Use the lastValue value to create a new query condition that matches documents added after the
lastValue. Explicitly ensure $natural order with the hint() method:
query = QUERY( "insertDate" << GT << lastValue ).hint( BSON( "$natural" << 1 ) );

* Loop through the outer while (1) loop to re-query with the new query condition and repeat.
See also:
Detailed blog post on tailable cursor12

2.3.10 Isolate Sequence of Operations

Overview
Write operations are atomic on the level of a single document: no single write operation can atomically affect more
than one document or more than one collection.
When a single write operation modifies multiple documents, the operation as a whole is not atomic, and other operations may interleave. The modification of a single document, or record, is always atomic, even if the write operation
modifies multiple sub-document within the single record.
No other operations are atomic; however, you can isolate a single write operation that affects multiple documents
using the isolation operator.
This document describes one method of updating documents only if the local copy of the document reflects the current
state of the document in the database. In addition the following methods provide a way to manage isolated sequences
of operations:
the findAndModify() provides an isolated query and modify operation.
Perform Two Phase Commits (page 69)
12 http://shtylman.com/post/the-tail-of-mongodb

76

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

Create a unique index (page 334), to ensure that a key doesnt exist when you insert it.
Update if Current
In this pattern, you will:
query for a document,
modify the fields in that document
and update the fields of a document only if the fields have not changed in the collection since the query.
Consider the following example in JavaScript which attempts to update the qty field of a document in the products
collection:
var myCollection = db.products;
var myDocument = myCollection.findOne( { sku: 'abc123' } );
if (myDocument) {
var oldQty = myDocument.qty;
if (myDocument.qty < 10) {
myDocument.qty *= 4;
} else if ( myDocument.qty < 20 ) {
myDocument.qty *= 3;
} else {
myDocument.qty *= 2;
}
myCollection.update(
{
_id: myDocument._id,
qty: oldQty
},
{
$set: { qty: myDocument.qty }
}
)
var err = db.getLastErrorObj();
if ( err && err.code ) {
print("unexpected error updating document: " + tojson( err ));
} else if ( err.n == 0 ) {
print("No update: no matching document for { _id: " + myDocument._id + ", qty: " + oldQty + " }")
}
}

Your application may require some modifications of this pattern, such as:
Use the entire document as the query in the update() operation, to generalize the operation and guarantee
that the original document was not modified, rather than ensuring that as single field was not changed.
Add a version variable to the document that applications increment upon each update operation to the documents.
Use this version variable in the query expression. You must be able to ensure that all clients that connect to your
database obey this constraint.
Use $set in the update expression to modify only your fields and prevent overriding other fields.

2.3. MongoDB CRUD Tutorials

77

MongoDB Documentation, Release 2.4.9

Use one of the methods described in Create an Auto-Incrementing Sequence Field (page 78).

2.3.11 Create an Auto-Incrementing Sequence Field

Synopsis
MongoDB reserves the _id field in the top level of all documents as a primary key. _id must be unique, and always
has an index with a unique constraint (page 334). However, except for the unique constraint you can use any value for
the _id field in your collections. This tutorial describes two methods for creating an incrementing sequence number
for the _id field using the following:
A Counters Collection (page 78)
Optimistic Loop (page 80)
Warning: Generally in MongoDB, you would not use an auto-increment pattern for the _id field, or any field,
because it does not scale for databases with large numbers of documents. Typically the default value ObjectId is
more ideal for the _id.

A Counters Collection

Use a separate counters collection to track the last number sequence used. The _id field contains the sequence
name and the seq field contains the last value of the sequence.
1. Insert into the counters collection, the initial value for the userid:
db.counters.insert(
{
_id: "userid",
seq: 0
}
)

2. Create a getNextSequence function that accepts a name of the sequence. The function uses the
findAndModify() method to atomically increment the seq value and return this new value:
function getNextSequence(name) {
var ret = db.counters.findAndModify(
{
query: { _id: name },
update: { $inc: { seq: 1 } },
new: true
}
);
return ret.seq;
}

3. Use this getNextSequence() function during insert().

db.users.insert(
{
_id: getNextSequence("userid"),
name: "Sarah C."
}
)

78

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

db.users.insert(
{
_id: getNextSequence("userid"),
name: "Bob D."
}
)

You can verify the results with find():

db.users.find()

The _id fields contain incrementing sequence values:

{
_id : 1,
name : "Sarah C."
}
{
_id : 2,
name : "Bob D."
}

Note: When findAndModify() includes the upsert: true option and the query field(s) is not uniquely
indexed, the method could insert a document multiple times in certain circumstances. For instance, if multiple clients
each invoke the method with the same query condition and these methods complete the find phase before any of
methods perform the modify phase, these methods could insert the same document.
In the counters collection example, the query field is the _id field, which always has a unique index. Consider
that the findAndModify() includes the upsert: true option, as in the following modified example:
function getNextSequence(name) {
var ret = db.counters.findAndModify(
{
query: { _id: name },
update: { $inc: { seq: 1 } },
new: true,
upsert: true
}
);
return ret.seq;
}

If multiple clients were to invoke the getNextSequence() method with the same name parameter, then the
methods would observe one of the following behaviors:
Exactly one findAndModify() would successfully insert a new document.
Zero or more findAndModify() methods would update the newly inserted document.
Zero or more findAndModify() methods would fail when they attempted to insert a duplicate.
If the method fails due to a unique index constraint violation, retry the method. Absent a delete of the document, the
retry should not fail.

2.3. MongoDB CRUD Tutorials

79

MongoDB Documentation, Release 2.4.9

Optimistic Loop

In this pattern, an Optimistic Loop calculates the incremented _id value and attempts to insert a document with the
calculated _id value. If the insert is successful, the loop ends. Otherwise, the loop will iterate through possible _id
values until the insert is successful.
1. Create a function named insertDocument that performs the insert if not present loop. The function wraps
the insert() method and takes a doc and a targetCollection arguments.
function insertDocument(doc, targetCollection) {
while (1) {
var cursor = targetCollection.find( {}, { _id: 1 } ).sort( { _id: -1 } ).limit(1);
var seq = cursor.hasNext() ? cursor.next()._id + 1 : 1;
doc._id = seq;
targetCollection.insert(doc);
var err = db.getLastErrorObj();
if( err && err.code ) {
if( err.code == 11000 /* dup key */ )
continue;
else
print( "unexpected error inserting data: " + tojson( err ) );
}
break;
}
}

The while (1) loop performs the following actions:

Queries the targetCollection for the document with the maximum _id value.
Determines the next sequence value for _id by:
adding 1 to the returned _id value if the returned cursor points to a document.
otherwise: it sets the next sequence value to 1 if the returned cursor points to no document.
For the doc to insert, set its _id field to the calculated sequence value seq.
Insert the doc into the targetCollection.
If the insert operation errors with duplicate key, repeat the loop. Otherwise, if the insert operation encounters some other error or if the operation succeeds, break out of the loop.
2. Use the insertDocument() function to perform an insert:
var myCollection = db.users2;
insertDocument(
{
name: "Grace H."
},
myCollection
);

80

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

insertDocument(
{
name: "Ted R."
},
myCollection
)

You can verify the results with find():

db.users2.find()

The _id fields contain incrementing sequence values:

{
_id: 1,
name: "Grace H."
}
{
_id : 2,
"name" : "Ted R."
}

The while loop may iterate many times in collections with larger insert volumes.

2.3.12 Limit Number of Elements in an Array after an Update

New in version 2.4.
Synopsis
Consider an application where users may submit many scores (e.g. for a test), but the application only needs to track
the top three test scores.
This pattern uses the $push operator with the $each, $sort, and $slice modifiers to sort and maintain an array
of fixed size.
Important: The array elements must be documents in order to use the $sort modifier.

Pattern
Consider the following document in the collection students:
{
_id: 1,
scores: [
{ attempt: 1, score: 10 },
{ attempt: 2 , score:8 }
]
}

The following update uses the $push operator with:

the $each modifier to append to the array 2 new elements,
the $sort modifier to order the elements by ascending (1) score, and

2.3. MongoDB CRUD Tutorials

81

MongoDB Documentation, Release 2.4.9

the $slice modifier to keep the last 3 elements of the ordered array.
db.students.update(
{ _id: 1 },
{ $push: { scores: { $each : [
{ attempt: 3, score: 7 },
{ attempt: 4, score: 4 }
],
$sort: { score: 1 },
$slice: -3
}
}
}
)

Note: When using the $sort modifier on the array element, access the field in the subdocument element directly
instead of using the dot notation on the array field.
After the operation, the document contains the only the top 3 scores in the scores array:
{
"_id" : 1,
"scores" : [
{ "attempt" : 3, "score" : 7 },
{ "attempt" : 2, "score" : 8 },
{ "attempt" : 1, "score" : 10 }
]
}

See also:
$push operator,
$each modifier,
$sort modifier, and
$slice modifier.

2.4 MongoDB CRUD Reference

2.4.1 Query Cursor Methods
Name
Description
cursor.count() Returns a count of the documents in a cursor.
cursor.explain() Reports on the query execution plan, including index use, for a cursor.
cursor.hint()
Forces MongoDB to use a specific index for a query.
cursor.limit() Constrains the size of a cursors result set.
cursor.next()
Returns the next document in a cursor.
cursor.skip()
Returns a cursor that begins returning results only after passing or skipping a number of
documents.
cursor.sort()
Returns results ordered according to a sort specification.
cursor.toArray() Returns an array that contains all documents returned by the cursor.

82

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

2.4.2 Query and Data Manipulation Collection Methods

Name
Description
db.collection.count() Wraps count to return a count of the number of documents in a collection or
matching a query.
db.collection.distinct()
Returns an array of documents that have distinct values for the specified field.
db.collection.find() Performs a query on a collection and returns a cursor object.
db.collection.findOne()Performs a query and returns a single document.
db.collection.insert() Creates a new document in a collection.
db.collection.remove() Deletes documents from a collection.
db.collection.save() Provides a wrapper around an insert() and update() to insert new
documents.
db.collection.update() Modifies a document in a collection.

2.4.3 MongoDB CRUD Reference Documentation

Write Concern Reference (page 83) Configuration options associated with the guarantee MongoDB provides when
reporting on the success of a write operation.
SQL to MongoDB Mapping Chart (page 85) An overview of common database operations showing both the MongoDB operations and SQL statements.
The bios Example Collection (page 90) Sample data for experimenting with MongoDB. insert(), update()
and find() pages use the data for some of their examples.
MongoDB Drivers and Client Libraries (page 95) Applications access MongoDB using client libraries, or drivers,
that provide idiomatic interfaces to MongoDB for many programming languages and development environments.
Write Concern Reference
Overview

Write concern describes the guarantee that MongoDB provides when reporting on the success of a write operation.
The strength of the write concerns determine the level of guarantee. When inserts, updates and deletes have a weak
write concern, write operations return quickly. In some failure cases, write operations issued with weak write concerns
may not persist. With stronger write concerns, clients wait after sending a write operation for MongoDB to confirm
the write operations.
MongoDB provides different levels of write concern to better address the specific needs of applications. Clients
may adjust write concern to ensure that the most important operations persist successfully to an entire MongoDB
deployment. For other less critical operations, clients can adjust the write concern to ensure faster performance rather
than ensure persistence to the entire deployment.
See also:
Write Concern (page 47) for an introduction to write concern in MongoDB.
Available Write Concern

To provide write concern, drivers (page 95) issue the getLastError command after a write operation and receive
a document with information about the last operation. This documents err field contains either:
null, which indicates the write operations have completed successfully, or

2.4. MongoDB CRUD Reference

83

MongoDB Documentation, Release 2.4.9

a description of the last error encountered.

The definition of a successful write depends on the arguments specified to getLastError, or in replica sets,
the configuration of getLastErrorDefaults (page 470). When deciding the level of write concern for your
application, see the introduction to Write Concern (page 47).
The getLastError command has the following options to configure write concern requirements:
j or journal option
This option confirms that the mongod instance has written the data to the on-disk journal and ensures data is
not lost if the mongod instance shuts down unexpectedly. Set to true to enable, as shown in the following
example:
db.runCommand( { getLastError: 1, j: "true" } )

If you set journal to true, and the mongod does not have journaling enabled, as with nojournal, then
getLastError will provide basic receipt acknowledgment, and will include a jnote field in its return
document.
w option
This option provides the ability to disable write concern entirely as well as specifies the write concern operations
for replica sets. See Write Concern Considerations (page 47) for an introduction to the fundamental concepts
of write concern. By default, the w option is set to 1, which provides basic receipt acknowledgment on a single
mongod instance or on the primary in a replica set.
The w option takes the following values:
-1:
Disables all acknowledgment of write operations, and suppresses all errors, including network and socket
errors.
0:
Disables basic acknowledgment of write operations, but returns information about socket exceptions and
networking errors to the application.
Note: If you disable basic write operation acknowledgment but require journal commit acknowledgment,
the journal commit prevails, and the driver will require that mongod will acknowledge the write operation.
1:
Provides acknowledgment of write operations on a standalone mongod or the primary in a replica set.
A number greater than 1:
Guarantees that write operations have propagated successfully to the specified number of replica set members including the primary. If you set w to a number that is greater than the number of set members that
hold data, MongoDB waits for the non-existent members to become available, which means MongoDB
blocks indefinitely.
majority:
Confirms that write operations have propagated to the majority of configured replica set: a majority of the
sets configured members must acknowledge the write operation before it succeeds. This allows you to
avoid hard coding assumptions about the size of your replica set into your application.
A tag set:
By specifying a tag set (page 450) you can have fine-grained control over which replica set members must
acknowledge a write operation to satisfy the required level of write concern.

84

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

getLastError also supports a wtimeout setting which allows clients to specify a timeout for the write concern:
if you dont specify wtimeout, or if you give it a value of 0, and the mongod cannot fulfill the write concern the
getLastError will block, potentially forever.
For more information on write concern and replica sets, see Write Concern for Replica Sets (page 49) for more
information.
In sharded clusters, mongos instances will pass write concern on to the shard mongod instances.
SQL to MongoDB Mapping Chart
In addition to the charts that follow, you might want to consider the Frequently Asked Questions (page 555) section for
a selection of common questions about MongoDB.
Terminology and Concepts

The following table presents the various SQL terminology and concepts and the corresponding MongoDB terminology
and concepts.
SQL Terms/Concepts
database
table
row
column
index
table joins
primary key
Specify any unique column or column combination as
primary key.
aggregation (e.g. group by)

MongoDB Terms/Concepts
database
collection
document or BSON document
field
index
embedded documents and linking
primary key
In MongoDB, the primary key is automatically set to
the _id field.
aggregation pipeline
See the SQL to Aggregation Mapping Chart
(page 310).

Executables

The following table presents the MySQL/Oracle executables and the corresponding MongoDB executables.
Database Server
Database Client

MySQL/Oracle
mysqld/oracle
mysql/sqlplus

MongoDB
mongod
mongo

Examples

The following table presents the various SQL statements and the corresponding MongoDB statements. The examples
in the table assume the following conditions:
The SQL examples assume a table named users.
The MongoDB examples assume a collection named users that contain documents of the following prototype:
{
_id: ObjectId("509a8fb2f3f4948bd2f983a0"),
user_id: "abc123",
age: 55,

2.4. MongoDB CRUD Reference

85

MongoDB Documentation, Release 2.4.9

status: 'A'
}

Create and Alter The following table presents the various SQL statements related to table-level actions and the
corresponding MongoDB statements.

86

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

SQL Schema Statements

CREATE TABLE users (
id MEDIUMINT NOT NULL
AUTO_INCREMENT,
user_id Varchar(30),
age Number,
status char(1),
PRIMARY KEY (id)
)

ALTER TABLE users

ADD join_date DATETIME

ALTER TABLE users

DROP COLUMN join_date

MongoDB Schema Statements

Reference
Implicitly
created
on
first See
insert()
and
insert() operation.
The pri- db.createCollection()
mary key _id is automatically added for more information.
if _id field is not specified.
db.users.insert( {
user_id: "abc123",
age: 55,
status: "A"
} )
However, you can also explicitly create a collection:
db.createCollection("users")
Collections do not describe or en- See the Data Modeling Concepts
force the structure of its documents; (page 99), update(), and $set
i.e. there is no structural alteration at for more information on changing the
the collection level.
structure of documents in a collecHowever, at the document level, tion.
update() operations can add fields
to existing documents using the
$set operator.
db.users.update(
{ },
{ $set: { join_date: new Date() } },
{ multi: true }
)
Collections do not describe or en- See Data Modeling Concepts
force the structure of its documents; (page 99), update(), and $unset
i.e. there is no structural alteration at for more information on changing
the collection level.
the structure of documents in a
However, at the document level, collection.
update() operations can remove
fields from documents using the
$unset operator.
db.users.update(
{ },
{ $unset: { join_date: "" } },
{ multi: true }
)

See ensureIndex() and indexes

CREATE INDEX idx_user_id_asc
db.users.ensureIndex( { user_id: 1 } )
(page 318) for more information.
ON users(user_id)
See ensureIndex() and indexes
CREATE INDEX
db.users.ensureIndex( { user_id: 1, age: -1 } )
(page 318) for more information.
idx_user_id_asc_age_desc
ON users(user_id, age DESC)
DROP TABLE users

2.4. MongoDB CRUD Reference

db.users.drop()

See drop() for more information.

87

MongoDB Documentation, Release 2.4.9

Insert The following table presents the various SQL statements related to inserting records into tables and the corresponding MongoDB statements.
SQL INSERT Statements

MongoDB insert() Statements

INSERT INTO users(user_id,

age,
status)
VALUES ("bcd001",
45,
"A")

db.users.insert( {
user_id: "bcd001",
age: 45,
status: "A"
} )

Reference
See insert() for more information.

Select The following table presents the various SQL statements related to reading records from tables and the corresponding MongoDB statements.

88

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

SQL SELECT Statements

MongoDB find() Statements

SELECT *
FROM users

db.users.find()

SELECT id, user_id, status db.users.find(

FROM users
{ },
{ user_id: 1, status: 1 }
)
SELECT user_id, status
FROM users

Reference
See find() for more information.

See find() for more information.

See find() for more information.

db.users.find(
{ },
{ user_id: 1, status: 1, _id: 0 }
)
See find() for more information.

SELECT *
FROM users
WHERE status = "A"

db.users.find(
{ status: "A" }
)

SELECT user_id, status

FROM users
WHERE status = "A"

See find() for more information.

db.users.find(
{ status: "A" },
{ user_id: 1, status: 1, _id: 0 }
)

SELECT *
FROM users
WHERE status != "A"

db.users.find(
{ status: { $ne: "A" } }
)

SELECT *
FROM users
WHERE status = "A"
AND age = 50

db.users.find(
{ status: "A",
age: 50 }
)

SELECT *
FROM users
WHERE status = "A"
OR age = 50

See find() and $or for more indb.users.find(

formation.
{ $or: [ { status: "A" } ,
{ age: 50 } ] }
)

SELECT *
FROM users
WHERE age > 25

db.users.find(
{ age: { $gt: 25 } }
)

SELECT *
FROM users
WHERE age < 25

db.users.find(
{ age: { $lt: 25 } }
)

SELECT *
FROM users
WHERE age > 25
AND
age <= 50
2.4.
MongoDB CRUD Reference
SELECT
*
FROM users
WHERE user_id like "%bc%"

See find() and $ne for more information.

See find() and $and for more information.

See find() and $gt for more information.

See find() and $lt for more information.

See find(), $gt, and $lte for

db.users.find(
more information.
{ age: { $gt: 25, $lte: 50 } }
)

db.users.find(
{ user_id: /bc/ }
)

See find() and $regex for more

89
information.

MongoDB Documentation, Release 2.4.9

Update Records The following table presents the various SQL statements related to updating existing records in
tables and the corresponding MongoDB statements.
SQL Update Statements

MongoDB update() Statements

UPDATE users
SET status = "C"
WHERE age > 25

db.users.update(
{ age: { $gt: 25 } },
{ $set: { status: "C" } },
{ multi: true }
)

UPDATE users
SET age = age + 3
WHERE status = "A"

db.users.update(
{ status: "A" } ,
{ $inc: { age: 3 } },
{ multi: true }
)

Reference
See update(), $gt, and $set for
more information.

See update(), $inc, and $set

for more information.

Delete Records The following table presents the various SQL statements related to deleting records from tables and
the corresponding MongoDB statements.
SQL Delete Statements
DELETE FROM users
WHERE status = "D"
DELETE FROM users

MongoDB remove() Statements

Reference
See remove() for more informadb.users.remove( { status: "D" } )
tion.

db.users.remove( )

See remove() for more information.

The bios Example Collection

The bios collection provides example data for experimenting with MongoDB. Many of this guides examples on
insert, update and read operations create or query data from the bios collection.
The following documents comprise the bios collection. In the examples, the data might be different, as the examples
themselves make changes to the data.
{
"_id" : 1,
"name" : {
"first" : "John",
"last" : "Backus"
},
"birth" : ISODate("1924-12-03T05:00:00Z"),
"death" : ISODate("2007-03-17T04:00:00Z"),
"contribs" : [
"Fortran",
"ALGOL",
"Backus-Naur Form",
"FP"
],
"awards" : [
{
"award" : "W.W. McDowell Award",
"year" : 1967,
"by" : "IEEE Computer Society"

90

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

},
{
"award" : "National Medal of Science",
"year" : 1975,
"by" : "National Science Foundation"
},
{
"award" : "Turing Award",
"year" : 1977,
"by" : "ACM"
},
{
"award" : "Draper Prize",
"year" : 1993,
"by" : "National Academy of Engineering"
}
]
}
{
"_id" : ObjectId("51df07b094c6acd67e492f41"),
"name" : {
"first" : "John",
"last" : "McCarthy"
},
"birth" : ISODate("1927-09-04T04:00:00Z"),
"death" : ISODate("2011-12-24T05:00:00Z"),
"contribs" : [
"Lisp",
"Artificial Intelligence",
"ALGOL"
],
"awards" : [
{
"award" : "Turing Award",
"year" : 1971,
"by" : "ACM"
},
{
"award" : "Kyoto Prize",
"year" : 1988,
"by" : "Inamori Foundation"
},
{
"award" : "National Medal of Science",
"year" : 1990,
"by" : "National Science Foundation"
}
]
}
{
"_id" : 3,
"name" : {
"first" : "Grace",
"last" : "Hopper"
},
"title" : "Rear Admiral",

2.4. MongoDB CRUD Reference

91

MongoDB Documentation, Release 2.4.9

"birth" : ISODate("1906-12-09T05:00:00Z"),
"death" : ISODate("1992-01-01T05:00:00Z"),
"contribs" : [
"UNIVAC",
"compiler",
"FLOW-MATIC",
"COBOL"
],
"awards" : [
{
"award" : "Computer Sciences Man of the Year",
"year" : 1969,
"by" : "Data Processing Management Association"
},
{
"award" : "Distinguished Fellow",
"year" : 1973,
"by" : " British Computer Society"
},
{
"award" : "W. W. McDowell Award",
"year" : 1976,
"by" : "IEEE Computer Society"
},
{
"award" : "National Medal of Technology",
"year" : 1991,
"by" : "United States"
}
]
}
{
"_id" : 4,
"name" : {
"first" : "Kristen",
"last" : "Nygaard"
},
"birth" : ISODate("1926-08-27T04:00:00Z"),
"death" : ISODate("2002-08-10T04:00:00Z"),
"contribs" : [
"OOP",
"Simula"
],
"awards" : [
{
"award" : "Rosing Prize",
"year" : 1999,
"by" : "Norwegian Data Association"
},
{
"award" : "Turing Award",
"year" : 2001,
"by" : "ACM"
},
{
"award" : "IEEE John von Neumann Medal",
"year" : 2001,

92

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

"by" : "IEEE"
}
]
}
{
"_id" : 5,
"name" : {
"first" : "Ole-Johan",
"last" : "Dahl"
},
"birth" : ISODate("1931-10-12T04:00:00Z"),
"death" : ISODate("2002-06-29T04:00:00Z"),
"contribs" : [
"OOP",
"Simula"
],
"awards" : [
{
"award" : "Rosing Prize",
"year" : 1999,
"by" : "Norwegian Data Association"
},
{
"award" : "Turing Award",
"year" : 2001,
"by" : "ACM"
},
{
"award" : "IEEE John von Neumann Medal",
"year" : 2001,
"by" : "IEEE"
}
]
}
{
"_id" : 6,
"name" : {
"first" : "Guido",
"last" : "van Rossum"
},
"birth" : ISODate("1956-01-31T05:00:00Z"),
"contribs" : [
"Python"
],
"awards" : [
{
"award" : "Award for the Advancement of Free Software",
"year" : 2001,
"by" : "Free Software Foundation"
},
{
"award" : "NLUUG Award",
"year" : 2003,
"by" : "NLUUG"
}
]

2.4. MongoDB CRUD Reference

93

MongoDB Documentation, Release 2.4.9

}
{
"_id" : ObjectId("51e062189c6ae665454e301d"),
"name" : {
"first" : "Dennis",
"last" : "Ritchie"
},
"birth" : ISODate("1941-09-09T04:00:00Z"),
"death" : ISODate("2011-10-12T04:00:00Z"),
"contribs" : [
"UNIX",
"C"
],
"awards" : [
{
"award" : "Turing Award",
"year" : 1983,
"by" : "ACM"
},
{
"award" : "National Medal of Technology",
"year" : 1998,
"by" : "United States"
},
{
"award" : "Japan Prize",
"year" : 2011,
"by" : "The Japan Prize Foundation"
}
]
}
{
"_id" : 8,
"name" : {
"first" : "Yukihiro",
"aka" : "Matz",
"last" : "Matsumoto"
},
"birth" : ISODate("1965-04-14T04:00:00Z"),
"contribs" : [
"Ruby"
],
"awards" : [
{
"award" : "Award for the Advancement of Free Software",
"year" : "2011",
"by" : "Free Software Foundation"
}
]
}
{
"_id" : 9,
"name" : {
"first" : "James",
"last" : "Gosling"

94

Chapter 2. MongoDB CRUD Operations

MongoDB Documentation, Release 2.4.9

},
"birth" : ISODate("1955-05-19T04:00:00Z"),
"contribs" : [
"Java"
],
"awards" : [
{
"award" : "The Economist Innovation Award",
"year" : 2002,
"by" : "The Economist"
},
{
"award" : "Officer of the Order of Canada",
"year" : 2007,
"by" : "Canada"
}
]
}
{
"_id" : 10,
"name" : {
"first" : "Martin",
"last" : "Odersky"
},
"contribs" : [
"Scala"
]
}

MongoDB Drivers and Client Libraries

An application communicates with MongoDB by way of a client library, called a driver13 , that handles all interaction
with the database in a language appropriate to the application.
Drivers

See the following pages for more information about the MongoDB drivers14 :
JavaScript (Language Center15 , docs16 )
Python (Language Center17 , docs18 )
Ruby (Language Center19 , docs20 )
PHP (Language Center21 , docs22 )
13 http://docs.mongodb.org/ecosystem/drivers
14 http://docs.mongodb.org/ecosystem/drivers
15 http://docs.mongodb.org/ecosystem/drivers/javascript
16 http://api.mongodb.org/js/current
17 http://docs.mongodb.org/ecosystem/drivers/python
18 http://api.mongodb.org/python/current
19 http://docs.mongodb.org/ecosystem/drivers/ruby
20 http://api.mongodb.org/ruby/current
21 http://docs.mongodb.org/ecosystem/drivers/php
22 http://php.net/mongo/

2.4. MongoDB CRUD Reference

95

MongoDB Documentation, Release 2.4.9

Perl (Language Center23 , docs24 )

Java (Language Center25 , docs26 )
Scala (Language Center27 , docs28 )
C# (Language Center29 , docs30 )
C (Language Center31 , docs32 )
C++ (Language Center33 , docs34 )
Haskell (Language Center35 , docs36 )
Erlang (Language Center37 , docs38 )
Driver Version Numbers

Driver version numbers use semantic versioning39 or major.minor.patch versioning system. The first number is the
major version, the second the minor version, and the third indicates a patch.
Example
Driver version numbers.
If your driver has a version number of 2.9.1, 2 is the major version, 9 is minor, and 1 is the patch.
The numbering scheme for drivers differs from the scheme for the MongoDB server. For more information on server
versioning, see MongoDB Version Numbers (page 634).

23 http://docs.mongodb.org/ecosystem/drivers/perl
24 http://api.mongodb.org/perl/current/
25 http://docs.mongodb.org/ecosystem/drivers/java
26 http://api.mongodb.org/java/current
27 http://docs.mongodb.org/ecosystem/drivers/scala
28 http://api.mongodb.org/scala/casbah/current/
29 http://docs.mongodb.org/ecosystem/drivers/csharp
30 http://api.mongodb.org/csharp/current/
31 http://docs.mongodb.org/ecosystem/drivers/c
32 http://api.mongodb.org/c/current/
33 http://docs.mongodb.org/ecosystem/drivers/cpp
34 http://api.mongodb.org/cplusplus/current/
35 http://hackage.haskell.org/package/mongoDB
36 http://api.mongodb.org/haskell/mongodb
37 http://docs.mongodb.org/ecosystem/drivers/erlang
38 http://api.mongodb.org/erlang/mongodb
39 http://semver.org/

96

Chapter 2. MongoDB CRUD Operations

CHAPTER 3

Data Models

Data in MongoDB has a flexible schema. Collections do not enforce document structure. This flexibility gives you
data-modeling choices to match your application and its performance requirements.
Read the Data Modeling Introduction (page 97) document for a high level introduction to data modeling, and proceed
to the documents in the Data Modeling Concepts (page 99) section for additional documentation of the data model
design process. The Data Model Examples and Patterns (page 106) documents provide examples of different data
models. In addition, the MongoDB Use Case Studies1 provide overviews of application design and include example
data models with MongoDB.
Data Modeling Introduction (page 97) An introduction to data modeling in MongoDB.
Data Modeling Concepts (page 99) The core documentation detailing the decisions you must make when determining a data model, and discussing considerations that should be taken into account.
Data Model Examples and Patterns (page 106) Examples of possible data models that you can use to structure your
MongoDB documents.
Data Model Reference (page 121) Reference material for data modeling for developers of MongoDB applications.

3.1 Data Modeling Introduction

Data in MongoDB has a flexible schema. Unlike SQL databases, where you must determine and declare a tables
schema before inserting data, MongoDBs collections do not enforce document structure. This flexibility facilitates
the mapping of documents to an entity or an object. Each document can match the data fields of the represented entity,
even if the data has substantial variation. In practice, however, the documents in a collection share a similar structure.
The key challenge in data modeling is balancing the needs of the application, the performance characteristics of the
database engine, and the data retrieval patterns. When designing data models, always consider the application usage
of the data (i.e. queries, updates, and processing of the data) as well as the inherent structure of the data itself.

3.1.1 Document Structure

The key decision in designing data models for MongoDB applications revolves around the structure of documents and
how the application represents relationships between data. There are two tools that allow applications to represent
these relationships: references and embedded documents.
1 http://docs.mongodb.org/ecosystem/use-cases

97

MongoDB Documentation, Release 2.4.9

References
References store the relationships between data by including links or references from one document to another. Applications can resolve these references (page 124) to access the related data. Broadly, these are normalized data models.

Figure 3.1: Data model using references to link documents. Both the contact document and the access document
contain a reference to the user document.
See Normalized Data Models (page 101) for the strengths and weaknesses of using references.
Embedded Data
Embedded documents capture relationships between data by storing related data in a single document structure. MongoDB documents make it possible to embed document structures as sub-documents in a field or array within a document. These denormalized data models allow applications to retrieve and manipulate related data in a single database
operation.
See Embedded Data Models (page 100) for the strengths and weaknesses of embedding sub-documents.

3.1.2 Atomicity of Write Operations

In MongoDB, write operations are atomic at the document level, and no single write operation can atomically affect
more than one document or more than one collection. A denormalized data model with embedded data combines
all related data for a represented entity in a single document. This facilitates atomic write operations since a single
write operation can insert or update the data for an entity. Normalizing the data would split the data across multiple
collections and would require multiple write operations that are not atomic collectively.

98

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

Figure 3.2: Data model with embedded fields that contain all related information.
However, schemas that facilitate atomic writes may limit ways that applications can use the data or may limit ways to
modify applications. The Atomicity Considerations (page 102) documentation describes the challenge of designing a
schema that balances flexibility and atomicity.

3.1.3 Document Growth

Some updates, such as pushing elements to an array or adding new fields, increase a documents size. If the document
size exceeds the allocated space for that document, MongoDB relocates the document on disk. The growth consideration can affect the decision to normalize or denormalize data. See Document Growth Considerations (page 102) for
more about planning for and managing document growth in MongoDB.

3.1.4 Data Use and Performance

When designing a data model, consider how applications will use your database. For instance, if your application only
uses recently inserted documents, consider using Capped Collections (page 161). Or if your application needs are
mainly read operations to a collection, adding indexes to support common queries can improve performance.
See Operational Factors and Data Models (page 102) for more information on these and other operational considerations that affect data model designs.

3.2 Data Modeling Concepts

When constructing a data model for your MongoDB collection, there are various options you can choose from, each
of which has its strengths and weaknesses. The following sections guide you through key design decisions and detail
various considerations for choosing the best data model for your application needs.

3.2. Data Modeling Concepts

99

MongoDB Documentation, Release 2.4.9

For a general introduction to data modeling in MongoDB, see the Data Modeling Introduction (page 97). For example
data models, see Data Modeling Examples and Patterns (page 106).
Data Model Design (page 100) Presents the different strategies that you can choose from when determining your data
model, their strengths and their weaknesses.
Operational Factors and Data Models (page 102) Details features you should keep in mind when designing your
data model, such as lifecycle management, indexing, horizontal scalability, and document growth.
GridFS (page 104) GridFS is a specification for storing documents that exceeds the BSON-document size limit of
16MB.

3.2.1 Data Model Design

Effective data models support your application needs. The key consideration for the structure of your documents is
the decision to embed (page 100) or to use references (page 101).
Embedded Data Models
With MongoDB, you may embed related data in a single structure or document. These schema are generally known
as denormalized models, and take advantage of MongoDBs rich documents. Consider the following diagram:

Figure 3.3: Data model with embedded fields that contain all related information.
Embedded data models allow applications to store related pieces of information in the same database record. As a
result, applications may need to issue fewer queries and updates to complete common operations.
In general, use embedded data models when:
you have contains relationships between entities. See Model One-to-One Relationships with Embedded Documents (page 106).

100

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

you have one-to-many relationships between entities. In these relationships the many or child documents
always appear with or are viewed in the context of the one or parent documents. See Model One-to-Many
Relationships with Embedded Documents (page 107).
In general, embedding provides better performance for read operations, as well as the ability to request and retrieve
related data in a single database operation. Embedded data models make it possible to update related data in a single
atomic write operation.
However, embedding related data in documents may lead to situations where documents grow after creation. Document growth can impact write performance and lead to data fragmentation. See Document Growth (page 102) for
details. Furthermore, documents in MongoDB must be smaller than the maximum BSON document size. For
bulk binary data, consider GridFS (page 104).
To interact with embedded documents, use dot notation to reach into embedded documents. See query for data
in arrays (page 62) and query data in sub-documents (page 61) for more examples on accessing data in arrays and
embedded documents.
Normalized Data Models
Normalized data models describe relationships using references (page 124) between documents.

Figure 3.4: Data model using references to link documents. Both the contact document and the access document
contain a reference to the user document.
In general, use normalized data models:
when embedding would result in duplication of data but would not provide sufficient read performance advantages to outweigh the implications of the duplication.
to represent more complex many-to-many relationships.

3.2. Data Modeling Concepts

101

MongoDB Documentation, Release 2.4.9

to model large hierarchical data sets.

References provides more flexibility than embedding. However, client-side applications must issue follow-up queries
to resolve the references. In other words, normalized data models can require more roundtrips to the server.
See Model One-to-Many Relationships with Document References (page 108) for an example of referencing. For
examples of various tree models using references, see Model Tree Structures (page 110).

3.2.2 Operational Factors and Data Models

Modeling application data for MongoDB depends on both the data itself, as well as the characteristics of MongoDB
itself. For example, different data models may allow applications to use more efficient queries, increase the throughput
of insert and update operations, or distribute activity to a sharded cluster more effectively.
These factors are operational or address requirements that arise outside of the application but impact the performance
of MongoDB based applications. When developing a data model, analyze all of your applications read operations
(page 31) and write operations (page 42) in conjunction with the following considerations.
Document Growth
Some updates to documents can increase the size of documents size. These updates include pushing elements to an
array (i.e. $push) and adding new fields to a document. If the document size exceeds the allocated space for that
document, MongoDB will relocate the document on disk. Relocating documents takes longer than in place updates and
can lead to fragmented storage. Although MongoDB automatically adds padding to document allocations (page 57)
to minimize the likelihood of relocation, data models should avoid document growth when possible.
For instance, if your applications require updates that will cause document growth, you may want to refactor your data
model to use references between data in distinct documents rather than a denormalized data model.
MongoDB adaptively adjusts the amount of automatic padding to reduce occurrences of relocation. You may also use
a pre-allocation strategy to explicitly avoid document growth. Refer to the Pre-Aggregated Reports Use Case2 for an
example of the pre-allocation approach to handling document growth.
Atomicity
In MongoDB, operations are atomic at the document level. No single write operation can change more than one
document. Operations that modify more than a single document in a collection still operate on one document at a time.
3
Ensure that your application stores all fields with atomic dependency requirements in the same document. If the
application can tolerate non-atomic updates for two pieces of data, you can store these data in separate documents.
A data model that embeds related data in a single document facilitates these kinds of atomic operations. For data models that store references between related pieces of data, the application must issue separate read and write operations
to retrieve and modify these related pieces of data.
See Model Data for Atomic Operations (page 119) for an example data model that provides atomic updates for a single
document.
Sharding
MongoDB uses sharding to provide horizontal scaling. These clusters support deployments with large data sets and
high-throughput operations. Sharding allows users to partition a collection within a database to distribute the collections documents across a number of mongod instances or shards.
2 http://docs.mongodb.org/ecosystem/use-cases/pre-aggregated-reports
3

Document-level atomic operations include all operations within a single MongoDB document record: operations that affect multiple subdocuments within that single record are still atomic.

102

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

To distribute data and application traffic in a sharded collection, MongoDB uses the shard key (page 492). Selecting
the proper shard key (page 492) has significant implications for performance, and can enable or prevent query isolation
and increased write capacity. It is important to consider carefully the field or fields to use as the shard key.
See Sharding Introduction (page 479) and Shard Keys (page 492) for more information.
Indexes
Use indexes to improve performance for common queries. Build indexes on fields that appear often in queries and for
all operations that return sorted results. MongoDB automatically creates a unique index on the _id field.
As you create indexes, consider the following behaviors of indexes:
Each index requires at least 8KB of data space.
Adding an index has some negative performance impact for write operations. For collections with high writeto-read ratio, indexes are expensive since each insert must also update any indexes.
Collections with high read-to-write ratio often benefit from additional indexes. Indexes do not affect un-indexed
read operations.
When active, each index consumes disk space and memory. This usage can be significant and should be tracked
for capacity planning, especially for concerns over working set size.
See Indexing Strategies (page 368) for more information on indexes as well as Analyze Query Performance (page 66).
Additionally, the MongoDB database profiler (page 174) may help identify inefficient queries.
Large Number of Collections
In certain situations, you might choose to store related information in several collections rather than in a single collection.
Consider a sample collection logs that stores log documents for various environment and applications. The logs
collection contains documents of the following form:
{ log: "dev", ts: ..., info: ... }
{ log: "debug", ts: ..., info: ...}

If the total number of documents is low, you may group documents into collection by type. For logs, consider maintaining distinct log collections, such as logs.dev and logs.debug. The logs.dev collection would contain
only the documents related to the dev environment.
Generally, having a large number of collections has no significant performance penalty and results in very good
performance. Distinct collections are very important for high-throughput batch processing.
When using models that have a large number of collections, consider the following behaviors:
Each collection has a certain minimum overhead of a few kilobytes.
Each index, including the index on _id, requires at least 8KB of data space.
For each database, a single namespace file (i.e. <database>.ns) stores all meta-data for that database, and
each index and collection has its own entry in the namespace file. MongoDB places limits on the size
of namespace files.
MongoDB has limits on the number of namespaces. You may wish to know the current number
of namespaces in order to determine how many additional namespaces the database can support. To get the
current number of namespaces, run the following in the mongo shell:

3.2. Data Modeling Concepts

103

MongoDB Documentation, Release 2.4.9

db.system.namespaces.count()

The limit on the number of namespaces depend on the <database>.ns size. The namespace file defaults to
16 MB.
To change the size of the new namespace file, start the server with the option --nssize <new size MB>.
For existing databases, after starting up the server with --nssize, run the db.repairDatabase() command from the mongo shell. For impacts and considerations on running db.repairDatabase(), see
repairDatabase.
Data Lifecycle Management
Data modeling decisions should take data lifecycle management into consideration.
The Time to Live or TTL feature (page 163) of collections expires documents after a period of time. Consider using
the TTL feature if your application requires some data to persist in the database for a limited period of time.
Additionally, if your application only uses recently inserted documents, consider Capped Collections (page 161).
Capped collections provide first-in-first-out (FIFO) management of inserted documents and efficiently support operations that insert and read documents based on insertion order.

3.2.3 GridFS
GridFS is a specification for storing and retrieving files that exceed the BSON-document size limit of 16MB.
Instead of storing a file in a single document, GridFS divides a file into parts, or chunks, 4 and stores each of those
chunks as a separate document. By default GridFS limits chunk size to 256k. GridFS uses two collections to store
files. One collection stores the file chunks, and the other stores file metadata.
When you query a GridFS store for a file, the driver or client will reassemble the chunks as needed. You can perform
range queries on files stored through GridFS. You also can access information from arbitrary sections of files, which
allows you to skip into the middle of a video or audio file.
GridFS is useful not only for storing files that exceed 16MB but also for storing any files for which you want access
without having to load the entire file into memory. For more information on the indications of GridFS, see When
should I use GridFS? (page 561).
Implement GridFS
To store and retrieve files using GridFS, use either of the following:
A MongoDB driver. See the drivers (page 95) documentation for information on using GridFS with your driver.
The
mongofiles
command-line
tool
in
the
mongo
shell.
http://docs.mongodb.org/manualreference/program/mongofiles.

See

GridFS Collections
GridFS stores files in two collections:
chunks stores the binary chunks. For details, see The chunks Collection (page 127).
files stores the files metadata. For details, see The files Collection (page 128).
4

The use of the term chunks in the context of GridFS is not related to the use of the term chunks in the context of sharding.

104

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

GridFS places the collections in a common bucket by prefixing each with the bucket name. By default, GridFS uses
two collections with names prefixed by fs bucket:
fs.files
fs.chunks
You can choose a different bucket name than fs, and create multiple buckets in a single database.
Each document in the chunks collection represents a distinct chunk of a file as represented in the GridFS store. Each
chunk is identified by its unique ObjectId stored in its _id field.
For descriptions of all fields in the chunks and files collections, see GridFS Reference (page 127).
GridFS Index
GridFS uses a unique, compound index on the chunks collection for the files_id and n fields. The files_id
field contains the _id of the chunks parent document. The n field contains the sequence number of the chunk.
GridFS numbers all chunks, starting with 0. For descriptions of the documents and fields in the chunks collection,
see GridFS Reference (page 127).
The GridFS index allows efficient retrieval of chunks using the files_id and n values, as shown in the following
example:
cursor = db.fs.chunks.find({files_id: myFileID}).sort({n:1});

See the relevant driver (page 95) documentation for the specific behavior of your GridFS application. If your driver
does not create this index, issue the following operation using the mongo shell:
db.fs.chunks.ensureIndex( { files_id: 1, n: 1 }, { unique: true } );

Example Interface
The following is an example of the GridFS interface in Java. The example is for demonstration purposes only. For
API specifics, see the relevant driver (page 95) documentation.
By default, the interface must support the default GridFS bucket, named fs, as in the following:
// returns default GridFS bucket (i.e. "fs" collection)
GridFS myFS = new GridFS(myDatabase);
// saves the file to "fs" GridFS bucket
myFS.createFile(new File("/tmp/largething.mpg"));

Optionally, interfaces may support other additional GridFS buckets as in the following example:
// returns GridFS bucket named "contracts"
GridFS myContracts = new GridFS(myDatabase, "contracts");
// retrieve GridFS object "smithco"
GridFSDBFile file = myContracts.findOne("smithco");
// saves the GridFS file to the file system
file.writeTo(new File("/tmp/smithco.pdf"));

3.2. Data Modeling Concepts

105

MongoDB Documentation, Release 2.4.9

3.3 Data Model Examples and Patterns

The following documents provide overviews of various data modeling patterns and common schema design considerations:
Model Relationships Between Documents (page 106) Examples for modeling relationships between documents.
Model One-to-One Relationships with Embedded Documents (page 106) Presents a data model that uses embedded documents (page 100) to describe one-to-one relationships between connected data.
Model One-to-Many Relationships with Embedded Documents (page 107) Presents a data model that uses
embedded documents (page 100) to describe one-to-many relationships between connected data.
Model One-to-Many Relationships with Document References (page 108) Presents a data model that uses
references (page 101) to describe one-to-many relationships between documents.
Model Tree Structures (page 110) Examples for modeling tree structures.
Model Tree Structures with Parent References (page 112) Presents a data model that organizes documents in
a tree-like structure by storing references (page 101) to parent nodes in child nodes.
Model Tree Structures with Child References (page 113) Presents a data model that organizes documents in a
tree-like structure by storing references (page 101) to child nodes in parent nodes.
See Model Tree Structures (page 110) for additional examples of data models for tree structures.
Model Specific Application Contexts (page 119) Examples for models for specific application contexts.
Model Data for Atomic Operations (page 119) Illustrates how embedding fields related to an atomic update
within the same document ensures that the fields are in sync
Model Data to Support Keyword Search (page 120) Describes one method for supporting keyword search by
storing keywords in an array in the same document as the text field. Combined with a multi-key index, this
pattern can support applications keyword search operations

3.3.1 Model Relationships Between Documents

Model One-to-One Relationships with Embedded Documents (page 106) Presents a data model that uses embedded
documents (page 100) to describe one-to-one relationships between connected data.
Model One-to-Many Relationships with Embedded Documents (page 107) Presents a data model that uses embedded documents (page 100) to describe one-to-many relationships between connected data.
Model One-to-Many Relationships with Document References (page 108) Presents a data model that uses references (page 101) to describe one-to-many relationships between documents.
Model One-to-One Relationships with Embedded Documents
Overview

Data in MongoDB has a flexible schema. Collections do not enforce document structure. Decisions that affect how
you model data can affect application performance and database capacity. See Data Modeling Concepts (page 99) for
a full high level overview of data modeling in MongoDB.
This document describes a data model that uses embedded (page 100) documents to describe relationships between
connected data.

106

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

Pattern

Consider the following example that maps patron and address relationships. The example illustrates the advantage of
embedding over referencing if you need to view one data entity in context of the other. In this one-to-one relationship
between patron and address data, the address belongs to the patron.
In the normalized data model, the address document contains a reference to the patron document.
{
_id: "joe",
name: "Joe Bookreader"
}
{
patron_id: "joe",
street: "123 Fake Street",
city: "Faketon",
state: "MA",
zip: 12345
}

If the address data is frequently retrieved with the name information, then with referencing, your application needs
to issue multiple queries to resolve the reference. The better data model would be to embed the address data in the
patron data, as in the following document:
{
_id: "joe",
name: "Joe Bookreader",
address: {
street: "123 Fake Street",
city: "Faketon",
state: "MA",
zip: 12345
}
}

With the embedded data model, your application can retrieve the complete patron information with one query.
Model One-to-Many Relationships with Embedded Documents
Overview

Data in MongoDB has a flexible schema. Collections do not enforce document structure. Decisions that affect how
you model data can affect application performance and database capacity. See Data Modeling Concepts (page 99) for
a full high level overview of data modeling in MongoDB.
This document describes a data model that uses embedded (page 100) documents to describe relationships between
connected data.
Pattern

Consider the following example that maps patron and multiple address relationships. The example illustrates the
advantage of embedding over referencing if you need to view many data entities in context of another. In this one-tomany relationship between patron and address data, the patron has multiple address entities.
In the normalized data model, the address documents contain a reference to the patron document.

3.3. Data Model Examples and Patterns

107

MongoDB Documentation, Release 2.4.9

{
_id: "joe",
name: "Joe Bookreader"
}
{
patron_id: "joe",
street: "123 Fake Street",
city: "Faketon",
state: "MA",
zip: 12345
}
{
patron_id: "joe",
street: "1 Some Other Street",
city: "Boston",
state: "MA",
zip: 12345
}

If your application frequently retrieves the address data with the name information, then your application needs
to issue multiple queries to resolve the references. A more optimal schema would be to embed the address data
entities in the patron data, as in the following document:
{
_id: "joe",
name: "Joe Bookreader",
addresses: [
{
street: "123 Fake Street",
city: "Faketon",
state: "MA",
zip: 12345
},
{
street: "1 Some Other Street",
city: "Boston",
state: "MA",
zip: 12345
}
]
}

With the embedded data model, your application can retrieve the complete patron information with one query.
Model One-to-Many Relationships with Document References
Overview

Data in MongoDB has a flexible schema. Collections do not enforce document structure. Decisions that affect how
you model data can affect application performance and database capacity. See Data Modeling Concepts (page 99) for
a full high level overview of data modeling in MongoDB.
This document describes a data model that uses references (page 101) between documents to describe relationships
between connected data.

108

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

Pattern

Consider the following example that maps publisher and book relationships. The example illustrates the advantage of
referencing over embedding to avoid repetition of the publisher information.
Embedding the publisher document inside the book document would lead to repetition of the publisher data, as the
following documents show:
{
title: "MongoDB: The Definitive Guide",
author: [ "Kristina Chodorow", "Mike Dirolf" ],
published_date: ISODate("2010-09-24"),
pages: 216,
language: "English",
publisher: {
name: "O'Reilly Media",
founded: 1980,
location: "CA"
}
}
{
title: "50 Tips and Tricks for MongoDB Developer",
author: "Kristina Chodorow",
published_date: ISODate("2011-05-06"),
pages: 68,
language: "English",
publisher: {
name: "O'Reilly Media",
founded: 1980,
location: "CA"
}
}

To avoid repetition of the publisher data, use references and keep the publisher information in a separate collection
from the book collection.
When using references, the growth of the relationships determine where to store the reference. If the number of books
per publisher is small with limited growth, storing the book reference inside the publisher document may sometimes
be useful. Otherwise, if the number of books per publisher is unbounded, this data model would lead to mutable,
growing arrays, as in the following example:
{
name: "O'Reilly Media",
founded: 1980,
location: "CA",
books: [12346789, 234567890, ...]
}
{
_id: 123456789,
title: "MongoDB: The Definitive Guide",
author: [ "Kristina Chodorow", "Mike Dirolf" ],
published_date: ISODate("2010-09-24"),
pages: 216,
language: "English"
}
{

3.3. Data Model Examples and Patterns

109

MongoDB Documentation, Release 2.4.9

_id: 234567890,
title: "50 Tips and Tricks for MongoDB Developer",
author: "Kristina Chodorow",
published_date: ISODate("2011-05-06"),
pages: 68,
language: "English"
}

To avoid mutable, growing arrays, store the publisher reference inside the book document:
{
_id: "oreilly",
name: "O'Reilly Media",
founded: 1980,
location: "CA"
}
{
_id: 123456789,
title: "MongoDB: The Definitive Guide",
author: [ "Kristina Chodorow", "Mike Dirolf" ],
published_date: ISODate("2010-09-24"),
pages: 216,
language: "English",
publisher_id: "oreilly"
}
{
_id: 234567890,
title: "50 Tips and Tricks for MongoDB Developer",
author: "Kristina Chodorow",
published_date: ISODate("2011-05-06"),
pages: 68,
language: "English",
publisher_id: "oreilly"
}

3.3.2 Model Tree Structures

MongoDB allows various ways to use tree data structures to model large hierarchical or nested data relationships.
Model Tree Structures with Parent References (page 112) Presents a data model that organizes documents in a treelike structure by storing references (page 101) to parent nodes in child nodes.
Model Tree Structures with Child References (page 113) Presents a data model that organizes documents in a treelike structure by storing references (page 101) to child nodes in parent nodes.
Model Tree Structures with an Array of Ancestors (page 115) Presents a data model that organizes documents in a
tree-like structure by storing references (page 101) to parent nodes and an array that stores all ancestors.
Model Tree Structures with Materialized Paths (page 116) Presents a data model that organizes documents in a treelike structure by storing full relationship paths between documents. In addition to the tree node, each document
stores the _id of the nodes ancestors or path as a string.
Model Tree Structures with Nested Sets (page 118) Presents a data model that organizes documents in a tree-like
structure using the Nested Sets pattern. This optimizes discovering subtrees at the expense of tree mutability.

110

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

Figure 3.5: Tree data model for a sample hierarchy of categories.

3.3. Data Model Examples and Patterns

111

MongoDB Documentation, Release 2.4.9

Model Tree Structures with Parent References

Overview

Data in MongoDB has a flexible schema. Collections do not enforce document structure. Decisions that affect how
you model data can affect application performance and database capacity. See Data Modeling Concepts (page 99) for
a full high level overview of data modeling in MongoDB.
This document describes a data model that describes a tree-like structure in MongoDB documents by storing references
(page 101) to parent nodes in children nodes.
Pattern

The Parent References pattern stores each tree node in a document; in addition to the tree node, the document stores
the id of the nodes parent.
Consider the following hierarchy of categories:

Figure 3.6: Tree data model for a sample hierarchy of categories.

The following example models the tree using Parent References, storing the reference to the parent category in the
field parent:
db.categories.insert( { _id: "MongoDB", parent: "Databases" } )
db.categories.insert( { _id: "dbm", parent: "Databases" } )
db.categories.insert( { _id: "Databases", parent: "Programming" } )

112

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

db.categories.insert( { _id: "Languages", parent: "Programming" } )

db.categories.insert( { _id: "Programming", parent: "Books" } )
db.categories.insert( { _id: "Books", parent: null } )

The query to retrieve the parent of a node is fast and straightforward:

db.categories.findOne( { _id: "MongoDB" } ).parent

You can create an index on the field parent to enable fast search by the parent node:
db.categories.ensureIndex( { parent: 1 } )

You can query by the parent field to find its immediate children nodes:
db.categories.find( { parent: "Databases" } )

The Parent Links pattern provides a simple solution to tree storage but requires multiple queries to retrieve subtrees.
Model Tree Structures with Child References
Overview

Data in MongoDB has a flexible schema. Collections do not enforce document structure. Decisions that affect how
you model data can affect application performance and database capacity. See Data Modeling Concepts (page 99) for
a full high level overview of data modeling in MongoDB.
This document describes a data model that describes a tree-like structure in MongoDB documents by storing references
(page 101) in the parent-nodes to children nodes.
Pattern

The Child References pattern stores each tree node in a document; in addition to the tree node, document stores in an
array the id(s) of the nodes children.
Consider the following hierarchy of categories:
The following example models the tree using Child References, storing the reference to the nodes children in the field
children:
db.categories.insert(
db.categories.insert(
db.categories.insert(
db.categories.insert(
db.categories.insert(
db.categories.insert(

{
{
{
{
{
{

_id:
_id:
_id:
_id:
_id:
_id:

"MongoDB", children: [] } )
"dbm", children: [] } )
"Databases", children: [ "MongoDB", "dbm" ] } )
"Languages", children: [] } )
"Programming", children: [ "Databases", "Languages" ] } )
"Books", children: [ "Programming" ] } )

The query to retrieve the immediate children of a node is fast and straightforward:
db.categories.findOne( { _id: "Databases" } ).children

You can create an index on the field children to enable fast search by the child nodes:
db.categories.ensureIndex( { children: 1 } )

You can query for a node in the children field to find its parent node as well as its siblings:
db.categories.find( { children: "MongoDB" } )

3.3. Data Model Examples and Patterns

113

MongoDB Documentation, Release 2.4.9

Figure 3.7: Tree data model for a sample hierarchy of categories.

114

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

The Child References pattern provides a suitable solution to tree storage as long as no operations on subtrees are
necessary. This pattern may also provide a suitable solution for storing graphs where a node may have multiple
parents.
Model Tree Structures with an Array of Ancestors
Overview

Data in MongoDB has a flexible schema. Collections do not enforce document structure. Decisions that affect how
you model data can affect application performance and database capacity. See Data Modeling Concepts (page 99) for
a full high level overview of data modeling in MongoDB.
This document describes a data model that describes a tree-like structure in MongoDB documents using references
(page 101) to parent nodes and an array that stores all ancestors.
Pattern

The Array of Ancestors pattern stores each tree node in a document; in addition to the tree node, document stores in
an array the id(s) of the nodes ancestors or path.
Consider the following hierarchy of categories:

Figure 3.8: Tree data model for a sample hierarchy of categories.

3.3. Data Model Examples and Patterns

115

MongoDB Documentation, Release 2.4.9

The following example models the tree using Array of Ancestors. In addition to the ancestors field, these documents also store the reference to the immediate parent category in the parent field:
db.categories.insert(
db.categories.insert(
db.categories.insert(
db.categories.insert(
db.categories.insert(
db.categories.insert(

{
{
{
{
{
{

_id:
_id:
_id:
_id:
_id:
_id:

"MongoDB", ancestors: [ "Books", "Programming", "Databases" ], parent: "

"dbm", ancestors: [ "Books", "Programming", "Databases" ], parent: "Data
"Databases", ancestors: [ "Books", "Programming" ], parent: "Programming
"Languages", ancestors: [ "Books", "Programming" ], parent: "Programming
"Programming", ancestors: [ "Books" ], parent: "Books" } )
"Books", ancestors: [ ], parent: null } )

The query to retrieve the ancestors or path of a node is fast and straightforward:
db.categories.findOne( { _id: "MongoDB" } ).ancestors

You can create an index on the field ancestors to enable fast search by the ancestors nodes:
db.categories.ensureIndex( { ancestors: 1 } )

You can query by the field ancestors to find all its descendants:
db.categories.find( { ancestors: "Programming" } )

The Array of Ancestors pattern provides a fast and efficient solution to find the descendants and the ancestors of a node
by creating an index on the elements of the ancestors field. This makes Array of Ancestors a good choice for working
with subtrees.
The Array of Ancestors pattern is slightly slower than the Materialized Paths (page 116) pattern but is more straightforward to use.
Model Tree Structures with Materialized Paths
Overview

Data in MongoDB has a flexible schema. Collections do not enforce document structure. Decisions that affect how
you model data can affect application performance and database capacity. See Data Modeling Concepts (page 99) for
a full high level overview of data modeling in MongoDB.
This document describes a data model that describes a tree-like structure in MongoDB documents by storing full
relationship paths between documents.
Pattern

The Materialized Paths pattern stores each tree node in a document; in addition to the tree node, document stores as
a string the id(s) of the nodes ancestors or path. Although the Materialized Paths pattern requires additional steps of
working with strings and regular expressions, the pattern also provides more flexibility in working with the path, such
as finding nodes by partial paths.
Consider the following hierarchy of categories:
The following example models the tree using Materialized Paths, storing the path in the field path; the path string
uses the comma , as a delimiter:
db.categories.insert(
db.categories.insert(
db.categories.insert(
db.categories.insert(
db.categories.insert(
db.categories.insert(

116

{
{
{
{
{
{

_id:
_id:
_id:
_id:
_id:
_id:

"Books", path: null } )

"Programming", path: ",Books," } )
"Databases", path: ",Books,Programming," } )
"Languages", path: ",Books,Programming," } )
"MongoDB", path: ",Books,Programming,Databases," } )
"dbm", path: ",Books,Programming,Databases," } )

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

Figure 3.9: Tree data model for a sample hierarchy of categories.

3.3. Data Model Examples and Patterns

117

MongoDB Documentation, Release 2.4.9

You can query to retrieve the whole tree, sorting by the field path:
db.categories.find().sort( { path: 1 } )

You can use regular expressions on the path field to find the descendants of Programming:
db.categories.find( { path: /,Programming,/ } )

You can also retrieve the descendants of Books where the Books is also at the topmost level of the hierarchy:
db.categories.find( { path: /^,Books,/ } )

To create an index on the field path use the following invocation:

db.categories.ensureIndex( { path: 1 } )

This index may improve performance depending on the query:

For queries of the Books sub-tree (e.g. http://docs.mongodb.org/manual^,Books,/) an
index on the path field improves the query performance significantly.
For queries of the Programming sub-tree (e.g. http://docs.mongodb.org/manual,Programming,/),
or similar queries of sub-tress, where the node might be in the middle of the indexed string, the query
must inspect the entire index.
For these queries an index may provide some performance improvement if the index is significantly smaller
than the entire collection.
Model Tree Structures with Nested Sets
Overview

Data in MongoDB has a flexible schema. Collections do not enforce document structure. Decisions that affect how
you model data can affect application performance and database capacity. See Data Modeling Concepts (page 99) for
a full high level overview of data modeling in MongoDB.
This document describes a data model that describes a tree like structure that optimizes discovering subtrees at the
expense of tree mutability.
Pattern

The Nested Sets pattern identifies each node in the tree as stops in a round-trip traversal of the tree. The application
visits each node in the tree twice; first during the initial trip, and second during the return trip. The Nested Sets pattern
stores each tree node in a document; in addition to the tree node, document stores the id of nodes parent, the nodes
initial stop in the left field, and its return stop in the right field.
Consider the following hierarchy of categories:
The following example models the tree using Nested Sets:
db.categories.insert(
db.categories.insert(
db.categories.insert(
db.categories.insert(
db.categories.insert(
db.categories.insert(

{
{
{
{
{
{

_id:
_id:
_id:
_id:
_id:
_id:

"Books", parent: 0, left: 1, right: 12 } )

"Programming", parent: "Books", left: 2, right: 11 } )
"Languages", parent: "Programming", left: 3, right: 4 } )
"Databases", parent: "Programming", left: 5, right: 10 } )
"MongoDB", parent: "Databases", left: 6, right: 7 } )
"dbm", parent: "Databases", left: 8, right: 9 } )

You can query to retrieve the descendants of a node:

118

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

Figure 3.10: Example of a hierarchical data. The numbers identify the stops at nodes during a roundtrip traversal of a
tree.

var databaseCategory = db.categories.findOne( { _id: "Databases" } );

db.categories.find( { left: { $gt: databaseCategory.left }, right: { $lt: databaseCategory.right } }

The Nested Sets pattern provides a fast and efficient solution for finding subtrees but is inefficient for modifying the
tree structure. As such, this pattern is best for static trees that do not change.

3.3.3 Model Specific Application Contexts

Model Data for Atomic Operations (page 119) Illustrates how embedding fields related to an atomic update within
the same document ensures that the fields are in sync
Model Data to Support Keyword Search (page 120) Describes one method for supporting keyword search by storing
keywords in an array in the same document as the text field. Combined with a multi-key index, this pattern can
support applications keyword search operations
Model Data for Atomic Operations
Pattern

Consider the following example that keeps a library book and its checkout information. The example illustrates how
embedding fields related to an atomic update within the same document ensures that the fields are in sync.
Consider the following book document that stores the number of available copies for checkout and the current checkout information:
book = {
_id: 123456789,
title: "MongoDB: The Definitive Guide",

3.3. Data Model Examples and Patterns

119

MongoDB Documentation, Release 2.4.9

author: [ "Kristina Chodorow", "Mike Dirolf" ],

published_date: ISODate("2010-09-24"),
pages: 216,
language: "English",
publisher_id: "oreilly",
available: 3,
checkout: [ { by: "joe", date: ISODate("2012-10-15") } ]
}

You can use the db.collection.findAndModify() method to atomically determine if a book is available for
checkout and update with the new checkout information. Embedding the available field and the checkout field
within the same document ensures that the updates to these fields are in sync:
db.books.findAndModify ( {
query: {
_id: 123456789,
available: { $gt: 0 }
},
update: {
$inc: { available: -1 },
$push: { checkout: { by: "abc", date: new Date() } }
}
} )

Model Data to Support Keyword Search

Note: Keyword search is not the same as text search or full text search, and does not provide stemming or other
text-processing features. See the Limitations of Keyword Indexes (page 121) section for more information.
In 2.4, MongoDB provides a text search feature. See Text Indexes (page 332) for more information.
If your application needs to perform queries on the content of a field that holds text you can perform exact matches
on the text or use $regex to use regular expression pattern matches. However, for many operations on text, these
methods do not satisfy application requirements.
This pattern describes one method for supporting keyword search using MongoDB to support application search
functionality, that uses keywords stored in an array in the same document as the text field. Combined with a multi-key
index (page 324), this pattern can support applications keyword search operations.
Pattern

To add structures to your document to support keyword-based queries, create an array field in your documents and add
the keywords as strings in the array. You can then create a multi-key index (page 324) on the array and create queries
that select values from the array.
Example
Given a collection of library volumes that you want to provide topic-based search. For each volume, you add the array
topics, and you add as many keywords as needed for a given volume.
For the Moby-Dick volume you might have the following document:
{ title : "Moby-Dick" ,
author : "Herman Melville" ,
published : 1851 ,
ISBN : 0451526996 ,

120

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

topics : [ "whaling" , "allegory" , "revenge" , "American" ,

"novel" , "nautical" , "voyage" , "Cape Cod" ]
}

You then create a multi-key index on the topics array:

db.volumes.ensureIndex( { topics: 1 } )

The multi-key index creates separate index entries for each keyword in the topics array. For example the index
contains one entry for whaling and another for allegory.
You then query based on the keywords. For example:
db.volumes.findOne( { topics : "voyage" }, { title: 1 } )

Note: An array with a large number of elements, such as one with several hundreds or thousands of keywords will
incur greater indexing costs on insertion.

Limitations of Keyword Indexes

MongoDB can support keyword searches using specific data models and multi-key indexes (page 324); however, these
keyword indexes are not sufficient or comparable to full-text products in the following respects:
Stemming. Keyword queries in MongoDB can not parse keywords for root or related words.
Synonyms. Keyword-based search features must provide support for synonym or related queries in the application layer.
Ranking. The keyword look ups described in this document do not provide a way to weight results.
Asynchronous Indexing. MongoDB builds indexes synchronously, which means that the indexes used for keyword indexes are always current and can operate in real-time. However, asynchronous bulk indexes may be
more efficient for some kinds of content and workloads.

3.4 Data Model Reference

Documents (page 122) MongoDB stores all data in documents, which are JSON-style data structures composed of
field-and-value pairs.
Database References (page 124) Discusses manual references and DBRefs, which MongoDB can use to represent
relationships between documents.
GridFS Reference (page 127) Convention for storing large files in a MongoDB Database.
ObjectId (page 129) A 12-byte BSON type that MongoDB uses as the default value for its documents _id field if
the _id field is not specified.
BSON Types (page 131) Outlines the unique BSON types used by MongoDB. See BSONspec.org5 for the complete
BSON specification.
5 http://bsonspec.org/

3.4. Data Model Reference

121

MongoDB Documentation, Release 2.4.9

3.4.1 Documents
MongoDB stores all data in documents, which are JSON-style data structures composed of field-and-value pairs:
{ "item": "pencil", "qty": 500, "type": "no.2" }

Most user-accessible data structures in MongoDB are documents, including:

All database records.
Query selectors (page 31), which define what records to select for read, update, and delete operations.
Update definitions (page 42), which define what fields to modify during an update.
Index specifications (page 318), which define what fields to index.
Data output by MongoDB for reporting and configuration, such as the output of the serverStatus and the
replica set configuration document (page 467).
Document Format
MongoDB stores documents on disk in the BSON serialization format. BSON is a binary representation of JSON
documents, though contains more data types than does JSON. For the BSON spec, see bsonspec.org6 . See also BSON
Types (page 131).
The mongo JavaScript shell and the MongoDB language drivers (page 95) translate between BSON and the languagespecific document representation.
Document Structure
MongoDB documents are composed of field-and-value pairs and have the following structure:
{
field1:
field2:
field3:
...
fieldN:
}

value1,
value2,
value3,
valueN

The value of a field can be any of the BSON data types (page 131), including other documents, arrays, and arrays of
documents. The following document contains values of varying types:
var mydoc = {
_id: ObjectId("5099803df3f4948bd2f98391"),
name: { first: "Alan", last: "Turing" },
birth: new Date('Jun 23, 1912'),
death: new Date('Jun 07, 1954'),
contribs: [ "Turing machine", "Turing test", "Turingery" ],
views : NumberLong(1250000)
}

The above fields have the following data types:

_id holds an ObjectId.
name holds a subdocument that contains the fields first and last.
6 http://bsonspec.org/

122

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

birth and death hold values of the Date type.

contribs holds an array of strings.
views holds a value of the NumberLong type.
Field Names
Field names are strings. Field names cannot contain null characters, dots (.) or dollar signs ($). See Dollar Sign
Operator Escaping (page 562) for an alternate approach.
BSON documents may have more than one field with the same name. Most MongoDB interfaces (page 95), however,
represent MongoDB with a structure (e.g. a hash table) that does not support duplicate field names. If you need to
manipulate documents that have more than one field with the same name, see the driver documentation (page 95) for
your driver.
Some documents created by internal MongoDB processes may have duplicate fields, but no MongoDB process will
ever add duplicate fields to an existing user document.
Document Limitations
Documents have the following attributes:
The maximum BSON document size is 16 megabytes.
The maximum document size helps ensure that a single document cannot use excessive amount of RAM or,
during transmission, excessive amount of bandwidth. To store documents larger than the maximum size, MongoDB provides the GridFS API. See mongofiles and the documentation for your driver (page 95) for more
information about GridFS.
Documents (page 122) have the following restrictions on field names:
The field name _id is reserved for use as a primary key; its value must be unique in the collection, is
immutable, and may be of any type other than an array.
The field names cannot start with the $ character.
The field names cannot contain the . character.
MongoDB does not make guarantees regarding the order of fields in a BSON document. Drivers and MongoDB
will reorder the fields of a documents upon insertion and following updates.
Most programming languages represent BSON documents with some form of mapping type. Comparisons
between mapping type objects typically, depend on order. As a result, the only way to ensure that two documents
have the same set of field and value pairs is to compare each field and value individually.
The _id Field
The _id field has the following behavior and constraints:
In documents, the _id field is always indexed for regular collections.
The _id field may contain values of any BSON data type (page 131), other than an array.
Warning: To ensure functioning replication, do not store values that are of the BSON regular expression
type in the _id field.
The following are common options for storing values for _id:

3.4. Data Model Reference

123

MongoDB Documentation, Release 2.4.9

Use an ObjectId (page 129).

Use a natural unique identifier, if available. This saves space and avoids an additional index.
Generate an auto-incrementing number. See Create an Auto-Incrementing Sequence Field (page 78).
Generate a UUID in your application code. For a more efficient storage of the UUID values in the collection
and in the _id index, store the UUID as a value of the BSON BinData type.
Index keys that are of the BinData type are more efficiently stored in the index if:
the binary subtype value is in the range of 0-7 or 128-135, and
the length of the byte array is: 0, 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 16, 20, 24, or 32.
Use your drivers BSON UUID facility to generate UUIDs. Be aware that driver implementations may implement UUID serialization and deserialization logic differently, which may not be fully compatible with other
drivers. See your driver documentation7 for information concerning UUID interoperability.
Note: Most MongoDB driver clients will include the _id field and generate an ObjectId before sending the insert
operation to MongoDB; however, if the client sends a document without an _id field, the mongod will add the _id
field and generate the ObjectId.

Dot Notation
MongoDB uses the dot notation to access the elements of an array and to access the fields of a subdocument.
To access an element of an array by the zero-based index position, concatenate the array name with the dot (.) and
zero-based index position, and enclose in quotes:
'<array>.<index>'

To access a field of a subdocument with dot-notation, concatenate the subdocument name with the dot (.) and the
field name, and enclose in quotes:
'<subdocument>.<field>'

See also:
Subdocuments (page 61) for dot notation examples with subdocuments.
Arrays (page 62) for dot notation examples with arrays.

3.4.2 Database References

MongoDB does not support joins. In MongoDB some data is denormalized, or stored with related data in documents to
remove the need for joins. However, in some cases it makes sense to store related information in separate documents,
typically in different collections or databases.
MongoDB applications use one of two methods for relating documents:
1. Manual references (page 125) where you save the _id field of one document in another document as a reference.
Then your application can run a second query to return the embedded data. These references are simple and
sufficient for most use cases.
2. DBRefs (page 126) are references from one document to another using the value of the first documents _id
field collection, and optional database name. To resolve DBRefs, your application must perform additional
7 http://api.mongodb.org/

124

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

queries to return the referenced documents. Many drivers (page 95) have helper methods that form the query
for the DBRef automatically. The drivers 8 do not automatically resolve DBRefs into documents.
Use a DBRef when you need to embed documents from multiple collections in documents from one collection.
DBRefs also provide a common format and type to represent these relationships among documents. The DBRef
format provides common semantics for representing links between documents if your database must interact
with multiple frameworks and tools.
Unless you have a compelling reason for using a DBRef, use manual references.
Manual References
Background

Manual references refers to the practice of including one documents _id field in another document. The application
can then issue a second query to resolve the referenced fields as needed.
Process

Consider the following operation to insert two documents, using the _id field of the first document as a reference in
the second document:
original_id = ObjectId()
db.places.insert({
"_id": original_id,
"name": "Broadway Center",
"url": "bc.example.net"
})
db.people.insert({
"name": "Erin",
"places_id": original_id,
"url": "bc.example.net/Erin"
})

Then, when a query returns the document from the people collection you can, if needed, make a second query for
the document referenced by the places_id field in the places collection.
Use

For nearly every case where you want to store a relationship between two documents, use manual references
(page 125). The references are simple to create and your application can resolve references as needed.
The only limitation of manual linking is that these references do not convey the database and collection name. If you
have documents in a single collection that relate to documents in more than one collection, you may need to consider
using DBRefs (page 126).
8

Some community supported drivers may have alternate behavior and may resolve a DBRef into a document automatically.

3.4. Data Model Reference

125

MongoDB Documentation, Release 2.4.9

DBRefs
Background

DBRefs are a convention for representing a document, rather than a specific reference type. They include the name of
the collection, and in some cases the database, in addition to the value from the _id field.
Format

DBRefs have the following fields:

$ref
The $ref field holds the name of the collection where the referenced document resides.
$id
The $id field contains the value of the _id field in the referenced document.
$db
Optional.
Contains the name of the database where the referenced document resides.
Only some drivers support $db references.
Example
DBRef document would resemble the following:
{ "$ref" : <value>, "$id" : <value>, "$db" : <value> }

Consider a document from a collection that stored a DBRef in a creator field:

{
"_id" : ObjectId("5126bbf64aed4daf9e2ab771"),
// .. application fields
"creator" : {
"$ref" : "creators",
"$id" : ObjectId("5126bc054aed4daf9e2ab772"),
"$db" : "users"
}
}

The DBRef in this example, points to a document in the creators collection of the users database that has
ObjectId("5126bc054aed4daf9e2ab772") in its _id field.
Note: The order of fields in the DBRef matters, and you must use the above sequence when using a DBRef.

Support

C++ The C++ driver contains no support for DBRefs. You can transverse references manually.
C# The C# driver provides access to DBRef objects with the MongoDBRef Class9 and supplies the FetchDBRef
Method10 for accessing these objects.
9 http://api.mongodb.org/csharp/current/html/46c356d3-ed06-a6f8-42fa-e0909ab64ce2.htm
10 http://api.mongodb.org/csharp/current/html/1b0b8f48-ba98-1367-0a7d-6e01c8df436f.htm

126

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

Java The DBRef11 class provides supports for DBRefs from Java.
JavaScript The mongo shells JavaScript interface provides a DBRef.
Perl The Perl driver contains no support for DBRefs. You can transverse references manually or use the MongoDBx::AutoDeref12 CPAN module.
PHP The PHP driver does support DBRefs, including the optional $db reference, through The MongoDBRef class13 .
Python The Python driver provides the DBRef class14 , and the dereference method15 for interacting with DBRefs.
Ruby The Ruby Driver supports DBRefs using the DBRef class16 and the deference method17 .
Use

In most cases you should use the manual reference (page 125) method for connecting two or more related documents.
However, if you need to reference documents from multiple collections, consider a DBRef.

3.4.3 GridFS Reference

GridFS stores files in two collections:
chunks stores the binary chunks. For details, see The chunks Collection (page 127).
files stores the files metadata. For details, see The files Collection (page 128).
GridFS places the collections in a common bucket by prefixing each with the bucket name. By default, GridFS uses
two collections with names prefixed by fs bucket:
fs.files
fs.chunks
You can choose a different bucket name than fs, and create multiple buckets in a single database.
See also:
GridFS (page 104) for more information about GridFS.
The chunks Collection
Each document in the chunks collection represents a distinct chunk of a file as represented in the GridFS store. The
following is a prototype document from the chunks collection.:
{
"_id" : <ObjectId>,
"files_id" : <ObjectId>,
"n" : <num>,
"data" : <binary>
}

A document from the chunks collection contains the following fields:

11 http://api.mongodb.org/java/current/com/mongodb/DBRef.html
12 http://search.cpan.org/dist/MongoDBx-AutoDeref/
13 http://www.php.net/manual/en/class.mongodbref.php/
14 http://api.mongodb.org/python/current/api/bson/dbref.html
15 http://api.mongodb.org//python/current/api/pymongo/database.html#pymongo.database.Database.dereference
16 http://api.mongodb.org//ruby/current/BSON/DBRef.html
17 http://api.mongodb.org//ruby/current/Mongo/DB.html#dereference

3.4. Data Model Reference

127

MongoDB Documentation, Release 2.4.9

chunks._id
The unique ObjectId of the chunk.
chunks.files_id
The _id of the parent document, as specified in the files collection.
chunks.n
The sequence number of the chunk. GridFS numbers all chunks, starting with 0.
chunks.data
The chunks payload as a BSON binary type.
The chunks collection uses a compound index on files_id and n, as described in GridFS Index (page 105).
The files Collection
Each document in the files collection represents a file in the GridFS store. Consider the following prototype of a
document in the files collection:
{
"_id" : <ObjectId>,
"length" : <num>,
"chunkSize" : <num>
"uploadDate" : <timestamp>
"md5" : <hash>
"filename" : <string>,
"contentType" : <string>,
"aliases" : <string array>,
"metadata" : <dataObject>,
}

Documents in the files collection contain some or all of the following fields. Applications may create additional
arbitrary fields:
files._id
The unique ID for this document. The _id is of the data type you chose for the original document. The default
type for MongoDB documents is BSON ObjectId.
files.length
The size of the document in bytes.
files.chunkSize
The size of each chunk. GridFS divides the document into chunks of the size specified here. The default size is
256 kilobytes.
files.uploadDate
The date the document was first stored by GridFS. This value has the Date type.
files.md5
An MD5 hash returned from the filemd5 API. This value has the String type.
files.filename
Optional. A human-readable name for the document.
files.contentType
Optional. A valid MIME type for the document.
files.aliases
Optional. An array of alias strings.

128

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

files.metadata
Optional. Any additional information you want to store.

3.4.4 ObjectId
Overview
ObjectId is a 12-byte BSON type, constructed using:
a 4-byte value representing the seconds since the Unix epoch,
a 3-byte machine identifier,
a 2-byte process id, and
a 3-byte counter, starting with a random value.
In MongoDB, documents stored in a collection require a unique _id field that acts as a primary key. Because ObjectIds
are small, most likely unique, and fast to generate, MongoDB uses ObjectIds as the default value for the _id field if
the _id field is not specified. MongoDB clients should add an _id field with a unique ObjectId. However, if a client
does not add an _id field, mongod will add an _id field that holds an ObjectId.
Using ObjectIds for the _id field provides the following additional benefits:
in the mongo shell, you can access the creation time of the ObjectId, using the getTimestamp() method.
sorting on an _id field that stores ObjectId values is roughly equivalent to sorting by creation time.
Important: The relationship between the order of ObjectId values and generation time is not strict within a
single second. If multiple systems, or multiple processes or threads on a single system generate values, within a
single second; ObjectId values do not represent a strict insertion order. Clock skew between clients can also
result in non-strict ordering even for values, because client drivers generate ObjectId values, not the mongod
process.
Also consider the Documents (page 122) section for related information on MongoDBs document orientation.
ObjectId()
The mongo shell provides the ObjectId() wrapper class to generate a new ObjectId, and to provide the following
helper attribute and methods:
str
The hexadecimal string value of the ObjectId() object.
getTimestamp()
Returns the timestamp portion of the ObjectId() object as a Date.
toString()
Returns the string representation of the ObjectId() object. The returned string literal has the
format ObjectId(...).
Changed in version 2.2: In previous versions toString() returns the value of the ObjectId as a
hexadecimal string.
valueOf()

3.4. Data Model Reference

129

MongoDB Documentation, Release 2.4.9

Returns the value of the ObjectId() object as a hexadecimal string. The returned string is the
str attribute.
Changed in version 2.2: In previous versions valueOf() returns the ObjectId() object.
Examples
Consider the following uses ObjectId() class in the mongo shell:
Generate a new ObjectId

To generate a new ObjectId, use the ObjectId() constructor with no argument:

x = ObjectId()

In this example, the value of x would be:

ObjectId("507f1f77bcf86cd799439011")

To generate a new ObjectId using the ObjectId() constructor with a unique hexadecimal string:
y = ObjectId("507f191e810c19729de860ea")

In this example, the value of y would be:

ObjectId("507f191e810c19729de860ea")

To return the timestamp of an ObjectId() object, use the getTimestamp() method as follows:
Convert an ObjectId into a Timestamp

To return the timestamp of an ObjectId() object, use the getTimestamp() method as follows:
ObjectId("507f191e810c19729de860ea").getTimestamp()

This operation will return the following Date object:

ISODate("2012-10-17T20:46:22Z")

Convert ObjectIds into Strings

Access the str attribute of an ObjectId() object, as follows:

ObjectId("507f191e810c19729de860ea").str

This operation will return the following hexadecimal string:

507f191e810c19729de860ea

To return the value of an ObjectId() object as a hexadecimal string, use the valueOf() method as follows:
ObjectId("507f191e810c19729de860ea").valueOf()

This operation returns the following output:

130

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

507f191e810c19729de860ea

To return the string representation of an ObjectId() object, use the toString() method as follows:
ObjectId("507f191e810c19729de860ea").toString()

This operation will return the following output:

ObjectId("507f191e810c19729de860ea")

3.4.5 BSON Types

BSON is a binary serialization format used to store documents and make remote procedure calls in MongoDB. The
BSON specification is located at bsonspec.org18 .
BSON supports the following data types as values in documents. Each data type has a corresponding number that can
be used with the $type operator to query documents by BSON type.
Type
Double
String
Object
Array
Binary data
Object id
Boolean
Date
Null
Regular Expression
JavaScript
Symbol
JavaScript (with scope)
32-bit integer
Timestamp
64-bit integer
Min key
Max key

Number
1
2
3
4
5
7
8
9
10
11
13
14
15
16
17
18
255
127

When comparing values of different BSON types, MongoDB uses the following comparison order, from lowest to
highest:
1. MinKey (internal type)
2. Null
3. Numbers (ints, longs, doubles)
4. Symbol, String
5. Object
6. Array
7. BinData
8. ObjectID
9. Boolean
18 http://bsonspec.org/

3.4. Data Model Reference

131

MongoDB Documentation, Release 2.4.9

10. Date, Timestamp

11. Regular Expression
12. MaxKey (internal type)
MongoDB treats some types as equivalent for comparison purposes. For instance, numeric types undergo conversion
before comparison.
The comparison treats a non-existent field as it would an empty BSON Object. As such, a sort on the a field in
documents { } and { a: null } would treat the documents as equivalent in sort order.
With arrays, a less-than comparison or an ascending sort compares the smallest element of arrays, and a greater-than
comparison or a descending sort compares the largest element of the arrays. As such, when comparing a field whose
value is a single-element array (e.g. [ 1 ]) with non-array fields (e.g. 2), the comparison is between 1 and 2. A
comparison of an empty array (e.g. [ ]) treats the empty array as less than null or a missing field.
To determine a fields type, see Check Types in the mongo Shell (page 209).
If you convert BSON to JSON, see Data Type Fidelity (page 151) for more information.
The next sections describe special considerations for particular BSON types.
ObjectId
ObjectIds are: small, likely unique, fast to generate, and ordered. These values consists of 12-bytes, where the first
four bytes are a timestamp that reflect the ObjectIds creation. Refer to the ObjectId (page 129) documentation for
more information.
String
BSON strings are UTF-8. In general, drivers for each programming language convert from the languages string format
to UTF-8 when serializing and deserializing BSON. This makes it possible to store most international characters in
BSON strings with ease. 19 In addition, MongoDB $regex queries support UTF-8 in the regex string.
Timestamps
BSON has a special timestamp type for internal MongoDB use and is not associated with the regular Date (page 133)
type. Timestamp values are a 64 bit value where:
the first 32 bits are a time_t value (seconds since the Unix epoch)
the second 32 bits are an incrementing ordinal for operations within a given second.
Within a single mongod instance, timestamp values are always unique.
In replication, the oplog has a ts field. The values in this field reflect the operation time, which uses a BSON
timestamp value.
Note: The BSON Timestamp type is for internal MongoDB use. For most cases, in application development, you
will want to use the BSON date type. See Date (page 133) for more information.
If you create a BSON Timestamp using the empty constructor (e.g. new Timestamp()), MongoDB will only
generate a timestamp if you use the constructor in the first field of the document. 20 Otherwise, MongoDB will
generate an empty timestamp value (i.e. Timestamp(0, 0).)
19 Given strings using UTF-8 character sets, using sort() on strings will be reasonably correct. However, because internally sort() uses the
C++ strcmp api, the sort order may handle some characters incorrectly.
20 If the first field in the document is _id, then you can generate a timestamp in the second field of a document.

132

Chapter 3. Data Models

MongoDB Documentation, Release 2.4.9

Changed in version 2.1: mongo shell displays the Timestamp value with the wrapper:
Timestamp(<time_t>, <ordinal>)

Prior to version 2.1, the mongo shell display the Timestamp value as a document:
{ t : <time_t>, i : <ordinal> }

Date
BSON Date is a 64-bit integer that represents the number of milliseconds since the Unix epoch (Jan 1, 1970). The
official BSON specification21 refers to the BSON Date type as the UTC datetime.
Changed in version 2.0: BSON Date type is signed.

22

Negative values represent dates before 1970.

Example
Construct a Date using the new Date() constructor in the mongo shell:
var mydate1 = new Date()

Example
Construct a Date using the ISODate() constructor in the mongo shell:
var mydate2 = ISODate()

Example
Return the Date value as string:
mydate1.toString()

Example
Return the month portion of the Date value; months are zero-indexed, so that January is month 0:
mydate1.getMonth()

21 http://bsonspec.org/#/specification
22 Prior to version 2.0, Date values were incorrectly interpreted as unsigned integers, which affected sorts, range queries, and indexes on Date
fields. Because indexes are not recreated when upgrading, please re-index if you created an index on Date values with an earlier version, and dates
before 1970 are relevant to your application.

3.4. Data Model Reference

133

MongoDB Documentation, Release 2.4.9

134

Chapter 3. Data Models

CHAPTER 4

Administration

The administration documentation addresses the ongoing operation and maintenance of MongoDB instances and deployments. This documentation includes both high level overviews of these concerns as well as tutorials that cover
specific procedures and processes for operating MongoDB.
Administration Concepts (page 135) Core conceptual documentation of operational practices for managing MongoDB deployments and systems.
MongoDB Backup Methods (page 136) Describes approaches and considerations for backing up a MongoDB
database.
Data Center Awareness (page 159) Presents the MongoDB features that allow application developers and
database administrators to configure their deployments to be more data center aware or allow operational
and location-based separation.
Monitoring for MongoDB (page 138) An overview of monitoring tools, diagnostic strategies, and approaches
to monitoring replica sets and sharded clusters.
Administration Tutorials (page 169) Tutorials that describe common administrative procedures and practices for operations for MongoDB instances and deployments.
Configuration, Maintenance, and Analysis (page 170) Describes routine management operations, including
configuration and performance analysis.
Backup and Recovery (page 190) Outlines procedures for data backup and restoration with mongod instances
and deployments.
Administration Reference (page 223) Reference and documentation of internal mechanics of administrative features,
systems and functions and operations.
See also:
The MongoDB Manual contains administrative documentation and tutorials though out several sections. See Replica
Set Tutorials (page 419) and Sharded Cluster Tutorials (page 506) for additional tutorials and information.

4.1 Administration Concepts

The core administration documents address strategies and practices used in the operation of MongoDB systems and
deployments.
Operational Strategies (page 136) Higher level documentation of key concepts for the operation and maintenance of
MongoDB deployments, including backup, maintenance, and configuration.
MongoDB Backup Methods (page 136) Describes approaches and considerations for backing up a MongoDB
database.
135

MongoDB Documentation, Release 2.4.9

Monitoring for MongoDB (page 138) An overview of monitoring tools, diagnostic strategies, and approaches
to monitoring replica sets and sharded clusters.
Run-time Database Configuration (page 146) Outlines common MongoDB configurations and examples of
best-practice configurations for common use cases.
Data Management (page 159) Core documentation that addresses issues in data management, organization, maintenance, and lifestyle management.
Data Center Awareness (page 159) Presents the MongoDB features that allow application developers and
database administrators to configure their deployments to be more data center aware or allow operational
and location-based separation.
Expire Data from Collections by Setting TTL (page 163) TTL collections make it possible to automatically
remove data from a collection based on the value of a timestamp and are useful for managing data like
machine generated event data that are only useful for a limited period of time.
Capped Collections (page 161) Capped collections provide a special type of size-constrained collections that
preserve insertion order and can support high volume inserts.
Optimization Strategies for MongoDB (page 165) Techniques for optimizing application performance with MongoDB.

4.1.1 Operational Strategies

These documents address higher level strategies for common administrative tasks and requirements with respect to
MongoDB deployments.
MongoDB Backup Methods (page 136) Describes approaches and considerations for backing up a MongoDB
database.
Monitoring for MongoDB (page 138) An overview of monitoring tools, diagnostic strategies, and approaches to
monitoring replica sets and sharded clusters.
Run-time Database Configuration (page 146) Outlines common MongoDB configurations and examples of bestpractice configurations for common use cases.
Import and Export MongoDB Data (page 150) Provides an overview of mongoimport and mongoexport, the
tools MongoDB includes for importing and exporting data.
Production Notes (page 153) A collection of notes that describe best practices and considerations for the operations
of MongoDB instances and deployments.
MongoDB Backup Methods
When deploying MongoDB in production, you should have a strategy for capturing and restoring backups in the case
of data loss events. MongoDB provides backup methods to support different requirements and configurations:
Backups with the MongoDB Management Service (MMS) (page 136)
Backup by Copying Underlying Data Files (page 137)
Backup with mongodump (page 137)
Backup Methods

136

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Backups with the MongoDB Management Service (MMS) The MongoDB Management Service1 supports backup
and restore for MongoDB deployments.
MMS continually backs up MongoDB replica sets and sharded systems by reading the oplog data from your MongoDB
cluster.
MMS Backup offers point in time recovery of MongoDB replica sets and a consistent snapshot of sharded systems.
MMS achieves point in time recovery by storing oplog data so that it can create a restore for any moment in time in
the last 24 hours for a particular replica set.
For sharded systems, MMS does not provide restores for arbitrary moments in time. MMS does provide periodic consistent snapshots of the entire sharded cluster. Sharded cluster snapshots are difficult to achieve with other MongoDB
backup methods.
To restore a MongoDB cluster from an MMS Backup snapshot, you download a compressed archive of your MongoDB
data files and distribute those files before restarting the mongod processes.
To get started with MMS Backup sign up for MMS2 , and consider the complete documentation of MMS see the MMS
Manual3 .
Backup by Copying Underlying Data Files You can create a backup by copying MongoDBs underlying data files.
If the volume where MongoDB stores data files supports point in time snapshots, you can use these snapshots to create
backups of a MongoDB system at an exact moment in time.
File systems snapshots are an operating system volume manager feature, and are not specific to MongoDB. The
mechanics of snapshots depend on the underlying storage system. For example, if you use Amazons EBS storage
system for EC2 supports snapshots. On Linux the LVM manager can create a snapshot.
To get a correct snapshot of a running mongod process, you must have journaling enabled and the journal must reside
on the same logical volume as the other MongoDB data files. Without journaling enabled, there is no guarantee that
the snapshot will be consistent or valid.
To get a consistent snapshot of a sharded system, you must disable the balancer and capture a snapshot from every
shard and a config server at approximately the same moment in time.
If your storage system does not support snapshots, you can copy the files directly using cp, rsync, or a similar tool.
Since copying multiple files is not an atomic operation, you must stop all writes to the mongod before copying the
files. Otherwise, you will copy the files in an invalid state.
Backups produced by copying the underlying data do not support point in time recovery for replica sets and are
difficult to manage for larger sharded clusters. Additionally, these backups are larger because they include the indexes
and duplicate underlying storage padding and fragmentation. mongodump by contrast create smaller backups.
For more information, see Backup and Restore with Filesystem Snapshots (page 190) and Backup a Sharded Cluster
with Filesystem Snapshots (page 199) documents for complete instructions on using LVM to create snapshots. Also
see Back up and Restore Processes for MongoDB on Amazon EC24 .
Backup with mongodump The mongodump tool reads data from a MongoDB database and creates high fidelity
BSON files. The mongorestore tool can populate a MongoDB database with the data from these BSON files.
These tools are simple and efficient for backing up small MongoDB deployments, but are not ideal for capturing
backups of larger systems.
mongodump and mongorestore can operate against a running mongod process, and can manipulate the underlying data files directly. By default, mongodump does not capture the contents of the local database (page 472).
1 https://mms.10gen.com/?pk_campaign=MongoDB-Org&pk_kwd=Backup-Docs
2 http://mms.mongodb.com
3 https://mms.mongodb.com/help/
4 http://docs.mongodb.org/ecosystem/tutorial/backup-and-restore-mongodb-on-amazon-ec2

4.1. Administration Concepts

137

MongoDB Documentation, Release 2.4.9

mongodump only captures the documents in the database. The resulting backup is space efficient, but
mongorestore or mongod must rebuild the indexes after restoring data.
When connected to a MongoDB instance, mongodump can adversely affect mongod performance. If your data is
larger than system memory, the queries will push the working set out of memory.
To mitigate the impact of mongodump on the performance of the replica set, use mongodump to capture backups from a secondary (page 382) member of a replica set. Alternatively, you can shut down a secondary and use
mongodump with the data files directly. If you shut down a secondary to capture data with mongodump ensure that
the operation can complete before its oplog becomes too stale to continue replicating.
For replica sets, mongodump also supports a point in time feature with the --oplog option. Applications may
continue modifying data while mongodump captures the output. To restore a point in time backup created with
--oplog, use mongorestore with the --oplogReplay option.
If applications modify data while mongodump is creating a backup, mongodump will compete for resources with
those applications.
See Back Up and Restore with MongoDB Tools (page 195), Backup a Small Sharded Cluster with mongodump
(page 198), and Backup a Sharded Cluster with Database Dumps (page 200) for more information.
Further Reading

Backup and Restore with Filesystem Snapshots (page 190) An outline of procedures for creating MongoDB data set
backups using system-level file snapshot tool, such as LVM or native storage appliance tools.
Restore a Replica Set from MongoDB Backups (page 194) Describes procedure for restoring a replica set from an
archived backup such as a mongodump or MMS Backup5 file.
Back Up and Restore with MongoDB Tools (page 195) The procedure for writing the contents of a database to a
BSON (i.e. binary) dump file for backing up MongoDB databases.
Backup and Restore Sharded Clusters (page 198) Detailed procedures and considerations for backing up sharded
clusters and single shards.
Recover Data after an Unexpected Shutdown (page 203) Recover data from MongoDB data files that were not properly closed or have an invalid state.
Monitoring for MongoDB
Monitoring is a critical component of all database administration. A firm grasp of MongoDBs reporting will allow you
to assess the state of your database and maintain your deployment without crisis. Additionally, a sense of MongoDBs
normal operational parameters will allow you to diagnose before they escalate to failures.
This document presents an overview of the available monitoring utilities and the reporting statistics available in MongoDB. It also introduces diagnostic strategies and suggestions for monitoring replica sets and sharded clusters.
Note: MongoDB Management Service (MMS)6 is a hosted monitoring service which collects and aggregates data
to provide insight into the performance and operation of MongoDB deployments. See the MMS documentation7 for
more information.
5 http://mms.mongodb.com
6 https://mms.mongodb.com/?pk_campaign=mongodb-org&pk_kwd=monitoring
7 http://mms.mongodb.com/help/

138

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Monitoring Strategies

There are three methods for collecting data about the state of a running MongoDB instance:
First, there is a set of utilities distributed with MongoDB that provides real-time reporting of database activities.
Second, database commands return statistics regarding the current database state with greater fidelity.
Third, MMS Monitoring Service8 collects data from running MongoDB deployments and provides visualization
and alerts based on that data. MMS is a free service provided by MongoDB.
Each strategy can help answer different questions and is useful in different contexts. These methods are complementary.
MongoDB Reporting Tools

This section provides an overview of the reporting methods distributed with MongoDB. It also offers examples of the
kinds of questions that each method is best suited to help you address.
Utilities The MongoDB distribution includes a number of utilities that quickly return statistics about instances
performance and activity. Typically, these are most useful for diagnosing issues and assessing normal operation.
mongostat mongostat captures and returns the counts of database operations by type (e.g. insert, query, update,
delete, etc.). These counts report on the load distribution on the server.
Use mongostat to understand the distribution of operation types and to inform capacity planning. See the
mongostat manual for details.
mongotop mongotop tracks and reports the current read and write activity of a MongoDB instance, and reports
these statistics on a per collection basis.
Use mongotop to check if your database activity and use match your expectations. See the mongotop manual
for details.
REST Interface MongoDB provides a simple REST interface that can be useful for configuring monitoring and
alert scripts, and for other administrative tasks.
To enable, configure mongod to use REST, either by starting mongod with the --rest option, or by setting the
rest setting to true in a configuration file.
For more information on using the REST Interface see, the Simple REST Interface9 documentation.
HTTP Console MongoDB provides a web interface that exposes diagnostic and monitoring information in a simple
web page. The web interface is accessible at localhost:<port>, where the <port> number is 1000 more than
the mongod port .
For example, if a locally running mongod is using the default port 27017, access the HTTP console at
http://localhost:28017.
8 https://mms.mongodb.com/?pk_campaign=mongodb-org&pk_kwd=monitoring
9 http://docs.mongodb.org/ecosystem/tools/http-interfaces

4.1. Administration Concepts

139

MongoDB Documentation, Release 2.4.9

Commands MongoDB includes a number of commands that report on the state of the database.
These data may provide a finer level of granularity than the utilities discussed above. Consider using their output
in scripts and programs to develop custom alerts, or to modify the behavior of your application in response to the
activity of your instance. The db.currentOp method is another useful tool for identifying the database instances
in-progress operations.
serverStatus The serverStatus command, or db.serverStatus() from the shell, returns a general
overview of the status of the database, detailing disk usage, memory use, connection, journaling, and index access.
The command returns quickly and does not impact MongoDB performance.
serverStatus outputs an account of the state of a MongoDB instance. This command is rarely run directly. In
most cases, the data is more meaningful when aggregated, as one would see with monitoring tools including MMS10 .
Nevertheless, all administrators should be familiar with the data provided by serverStatus.
dbStats The dbStats command, or db.stats() from the shell, returns a document that addresses storage use
and data volumes. The dbStats reflect the amount of storage used, the quantity of data contained in the database,
and object, collection, and index counters.
Use this data to monitor the state and storage capacity of a specific database. This output also allows you to compare
use between databases and to determine the average document size in a database.
collStats The collStats provides statistics that resemble dbStats on the collection level, including a count
of the objects in the collection, the size of the collection, the amount of disk space used by the collection, and information about its indexes.
replSetGetStatus The replSetGetStatus command (rs.status() from the shell) returns an
overview of your replica sets status. The replSetGetStatus document details the state and configuration of
the replica set and statistics about its members.
Use this data to ensure that replication is properly configured, and to check the connections between the current host
and the other members of the replica set.
Third Party Tools A number of third party monitoring tools have support for MongoDB, either directly, or through
their own plugins.
Self Hosted Monitoring Tools These are monitoring tools that you must install, configure and maintain on your
own servers. Most are open source.
10 http://mms.mongodb.com

140

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Tool
Ganglia24
Ganglia
Motop27
mtop28
Munin29
Munin
Munin
Nagios33
Zabbix35

Plugin
mongodb-ganglia25

Description
Python script to report operations per second, memory usage, btree statistics,
master/slave status and current connections.
26
gmond_python_modules
Parses
output from the serverStatus and replSetGetStatus
commands.
None
Realtime monitoring tool for MongoDB servers. Shows current operations
ordered by durations every second.
None
A top like tool.
mongo-munin30
Retrieves server statistics.
mongomon31
Retrieves collection statistics (sizes, index sizes, and each (configured) collection
count for one DB).
munin-plugins
Some additional munin plugins not in the main distribution.
Ubuntu PPA32
nagios-pluginA simple Nagios check script, written in Python.
mongodb34
mikoomiMonitors availability, resource utilization, health, performance and other
mongodb36
important metrics.

Also consider dex37 , an index and query analyzing tool for MongoDB that compares MongoDB log files and indexes
to make indexing recommendations.
As part of MongoDB Enterprise38 , you can run MMS On-Prem39 , which offers the features of MMS in a package that
runs within your infrastructure.
Hosted (SaaS) Monitoring Tools These are monitoring tools provided as a hosted service, usually through a paid
subscription.
11 http://sourceforge.net/apps/trac/ganglia/wiki
12 https://github.com/quiiver/mongodb-ganglia
13 https://github.com/ganglia/gmond_python_modules
14 https://github.com/tart/motop
15 https://github.com/beaufour/mtop
16 http://munin-monitoring.org/
17 https://github.com/erh/mongo-munin
18 https://github.com/pcdummy/mongomon
19 https://launchpad.net/

chris-lea/+archive/munin-plugins

20 http://www.nagios.org/
21 https://github.com/mzupan/nagios-plugin-mongodb
22 http://www.zabbix.com/
23 https://code.google.com/p/mikoomi/wiki/03
24 http://sourceforge.net/apps/trac/ganglia/wiki
25 https://github.com/quiiver/mongodb-ganglia
26 https://github.com/ganglia/gmond_python_modules
27 https://github.com/tart/motop
28 https://github.com/beaufour/mtop
29 http://munin-monitoring.org/
30 https://github.com/erh/mongo-munin
31 https://github.com/pcdummy/mongomon
32 https://launchpad.net/

chris-lea/+archive/munin-plugins

33 http://www.nagios.org/
34 https://github.com/mzupan/nagios-plugin-mongodb
35 http://www.zabbix.com/
36 https://code.google.com/p/mikoomi/wiki/03
37 https://github.com/mongolab/dex
38 http://www.mongodb.com/products/mongodb-enterprise
39 http://mms.mongodb.com

4.1. Administration Concepts

141

MongoDB Documentation, Release 2.4.9

Name
MongoDB
Management
Service47
Scout48
Server Density52

Notes
MMS is a cloud-based suite of services for managing MongoDB deployments. MMS
provides monitoring and backup functionality.
Several plugins, including MongoDB Monitoring49 , MongoDB Slow Queries50 , and
MongoDB Replica Set Monitoring51 .
Dashboard for MongoDB53 , MongoDB specific alerts, replication failover timeline and
iPhone, iPad and Android mobile apps.

Process Logging

During normal operation, mongod and mongos instances report a live account of all server activity and operations to
either standard output or a log file. The following runtime settings control these options.
quiet. Limits the amount of information written to the log or output.
verbose. Increases the amount of information written to the log or output.
You can also specify this as v (as in -v). For higher levels of verbosity, set multiple v, as in vvvv = True.
You can also change the verbosity of a running mongod or mongos instance with the setParameter command.
logpath. Enables logging to a file, rather than the standard output. You must specify the full path to the log
file when adjusting this setting.
logappend. Adds information to a log file instead of overwriting the file.
Note: You can specify these configuration operations as the command line arguments to mongod or mongos
For example:
mongod -v --logpath /var/log/mongodb/server1.log --logappend

Starts a mongod instance in verbose

/var/log/mongodb/server1.log/.

mode,

appending

data

to

the

log

file

at

The following database commands also affect logging:

getLog. Displays recent messages from the mongod process log.
logRotate. Rotates the log files for mongod processes only. See Rotate Log Files (page 180).
40 https://mms.mongodb.com/?pk_campaign=mongodb-org&pk_kwd=monitoring
41 http://scoutapp.com
42 https://scoutapp.com/plugin_urls/391-mongodb-monitoring
43 http://scoutapp.com/plugin_urls/291-mongodb-slow-queries
44 http://scoutapp.com/plugin_urls/2251-mongodb-replica-set-monitoring
45 http://www.serverdensity.com
46 http://www.serverdensity.com/mongodb-monitoring/
47 https://mms.mongodb.com/?pk_campaign=mongodb-org&pk_kwd=monitoring
48 http://scoutapp.com
49 https://scoutapp.com/plugin_urls/391-mongodb-monitoring
50 http://scoutapp.com/plugin_urls/291-mongodb-slow-queries
51 http://scoutapp.com/plugin_urls/2251-mongodb-replica-set-monitoring
52 http://www.serverdensity.com
53 http://www.serverdensity.com/mongodb-monitoring/

142

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Diagnosing Performance Issues

Degraded performance in MongoDB is typically a function of the relationship between the quantity of data stored
in the database, the amount of system RAM, the number of connections to the database, and the amount of time the
database spends in a locked state.
In some cases performance issues may be transient and related to traffic load, data access patterns, or the availability
of hardware on the host system for virtualized environments. Some users also experience performance limitations as a
result of inadequate or inappropriate indexing strategies, or as a consequence of poor schema design patterns. In other
situations, performance issues may indicate that the database may be operating at capacity and that it is time to add
additional capacity to the database.
The following are some causes of degraded performance in MongoDB.
Locks MongoDB uses a locking system to ensure data set validity. However, if certain operations are long-running,
or a queue forms, performance will slow as requests and operations wait for the lock. Lock-related slowdowns can
be intermittent. To see if the lock has been affecting your performance, look to the data in the globalLock section of
the serverStatus output. If globalLock.currentQueue.total is consistently high, then there is a chance
that a large number of requests are waiting for a lock. This indicates a possible concurrency issue that may be affecting
performance.
If globalLock.totalTime is high relative to uptime, the database has existed in a lock state for a significant
amount of time. If globalLock.ratio is also high, MongoDB has likely been processing a large number of
long running queries. Long queries are often the result of a number of factors: ineffective use of indexes, nonoptimal schema design, poor query structure, system architecture issues, or insufficient RAM resulting in page faults
(page 143) and disk reads.
Memory Usage MongoDB uses memory mapped files to store data. Given a data set of sufficient size, the MongoDB
process will allocate all available memory on the system for its use. While this is part of the design, and affords
MongoDB superior performance, the memory mapped files make it difficult to determine if the amount of RAM is
sufficient for the data set.
The memory usage statuses metrics of the serverStatus output can provide insight into MongoDBs memory use.
Check the resident memory use (i.e. mem.resident): if this exceeds the amount of system memory and there is a
significant amount of data on disk that isnt in RAM, you may have exceeded the capacity of your system.
You should also check the amount of mapped memory (i.e. mem.mapped.) If this value is greater than the amount of
system memory, some operations will require disk access page faults to read data from virtual memory and negatively
affect performance.
Page Faults A page fault occurs when MongoDB requires data not located in physical memory, and must read from
virtual memory. To check for page faults, see the extra_info.page_faults value in the serverStatus
output. This data is only available on Linux systems.
A single page fault completes quickly and is not problematic. However, in aggregate, large volumes of page faults
typically indicate that MongoDB is reading too much data from disk. In many situations, MongoDBs read locks will
yield after a page fault to allow other processes to read and avoid blocking while waiting for the next page to read
into memory. This approach improves concurrency, and also improves overall throughput in high volume systems.
Increasing the amount of RAM accessible to MongoDB may help reduce the number of page faults. If this is not
possible, you may want to consider deploying a sharded cluster and/or adding shards to your deployment to distribute
load among mongod instances.

4.1. Administration Concepts

143

MongoDB Documentation, Release 2.4.9

Number of Connections In some cases, the number of connections between the application layer (i.e. clients) and
the database can overwhelm the ability of the server to handle requests. This can produce performance irregularities.
The following fields in the serverStatus document can provide insight:
globalLock.activeClients contains a counter of the total number of clients with active operations in
progress or queued.
connections is a container for the following two fields:
current the total number of current clients that connect to the database instance.
available the total number of unused collections available for new clients.
Note: Unless constrained by system-wide limits MongoDB has a hard connection limit of 20,000 connections. You
can modify system limits using the ulimit command, or by editing your systems /etc/sysctl file.
If requests are high because there are numerous concurrent application requests, the database may have trouble keeping
up with demand. If this is the case, then you will need to increase the capacity of your deployment. For read-heavy
applications increase the size of your replica set and distribute read operations to secondary members. For write heavy
applications, deploy sharding and add one or more shards to a sharded cluster to distribute load among mongod
instances.
Spikes in the number of connections can also be the result of application or driver errors. All of the officially supported
MongoDB drivers implement connection pooling, which allows clients to use and reuse connections more efficiently.
Extremely high numbers of connections, particularly without corresponding workload is often indicative of a driver or
other configuration error.
Database Profiling MongoDBs Profiler is a database profiling system that can help identify inefficient queries
and operations.
The following profiling levels are available:
Level
0
1
2

Setting
Off. No profiling
On. Only includes slow operations
On. Includes all operations

Enable the profiler by setting the profile value using the following command in the mongo shell:
db.setProfilingLevel(1)

The slowms setting defines what constitutes a slow operation. To set the threshold above which the profiler considers operations slow (and thus, included in the level 1 profiling data), you can configure slowms at runtime as an
argument to the db.setProfilingLevel() operation.
See
The documentation of db.setProfilingLevel() for more information about this command.
By default, mongod records all slow queries to its log, as defined by slowms. Unlike log data, the data in
system.profile does not persist between mongod restarts.
Note: Because the database profiler can negatively impact performance, only enable profiling for strategic intervals
and as minimally as possible on production systems.
You may enable profiling on a per-mongod basis. This setting will not propagate across a replica set or sharded
cluster.

144

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

You can view the output of the profiler in the system.profile collection of your database by issuing the show
profile command in the mongo shell, or with the following operation:
db.system.profile.find( { millis : { $gt : 100 } } )

This returns all operations that lasted longer than 100 milliseconds. Ensure that the value specified here (100, in this
example) is above the slowms threshold.
See also:
Optimization Strategies for MongoDB (page 165) addresses strategies that may improve the performance of your
database queries and operations.
Replication and Monitoring

Beyond the basic monitoring requirements for any MongoDB instance, for replica sets, administrators must monitor
replication lag. Replication lag refers to the amount of time that it takes to copy (i.e. replicate) a write operation
on the primary to a secondary. Some small delay period may be acceptable, but two significant problems emerge as
replication lag grows:
First, operations that occurred during the period of lag are not replicated to one or more secondaries. If youre
using replication to ensure data persistence, exceptionally long delays may impact the integrity of your data set.
Second, if the replication lag exceeds the length of the operation log (oplog) then MongoDB will have to perform
an initial sync on the secondary, copying all data from the primary and rebuilding all indexes. This is uncommon
under normal circumstances, but if you configure the oplog to be smaller than the default, the issue can arise.
Note: The size of the oplog is only configurable during the first run using the --oplogSize argument to the
mongod command, or preferably, the oplogSize in the MongoDB configuration file. If you do not specify
this on the command line before running with the --replSet option, mongod will create a default sized
oplog.
By default, the oplog is 5 percent of total available disk space on 64-bit systems. For more information about
changing the oplog size, see the Change the Size of the Oplog (page 445)
For causes of replication lag, see Replication Lag (page 461).
Replication issues are most often the result of network connectivity issues between members, or the result of a primary
that does not have the resources to support application and replication traffic. To check the status of a replica, use the
replSetGetStatus or the following helper in the shell:
rs.status()

The http://docs.mongodb.org/manualreference/command/replSetGetStatus document provides a more in-depth overview view of this output. In general, watch the value of optimeDate, and pay particular
attention to the time difference between the primary and the secondary members.
Sharding and Monitoring

In most cases, the components of sharded clusters benefit from the same monitoring and analysis as all other MongoDB
instances. In addition, clusters require further monitoring to ensure that data is effectively distributed among nodes
and that sharding operations are functioning appropriately.
See also:
See the Sharding Concepts (page 484) documentation for more information.

4.1. Administration Concepts

145

MongoDB Documentation, Release 2.4.9

Config Servers The config database maintains a map identifying which documents are on which shards. The cluster
updates this map as chunks move between shards. When a configuration server becomes inaccessible, certain sharding
operations become unavailable, such as moving chunks and starting mongos instances. However, clusters remain
accessible from already-running mongos instances.
Because inaccessible configuration servers can seriously impact the availability of a sharded cluster, you should monitor your configuration servers to ensure that the cluster remains well balanced and that mongos instances can restart.
MMS Monitoring54 monitors config servers and can create notifications if a config server becomes inaccessible.
Balancing and Chunk Distribution The most effective sharded cluster deployments evenly balance chunks among
the shards. To facilitate this, MongoDB has a background balancer process that distributes data to ensure that chunks
are always optimally distributed among the shards.
Issue the db.printShardingStatus() or sh.status() command to the mongos by way of the mongo
shell. This returns an overview of the entire cluster including the database name, and a list of the chunks.
Stale Locks In nearly every case, all locks used by the balancer are automatically released when they become stale.
However, because any long lasting lock can block future balancing, its important to ensure that all locks are legitimate.
To check the lock status of the database, connect to a mongos instance using the mongo shell. Issue the following
command sequence to switch to the config database and display all outstanding locks on the shard database:
use config
db.locks.find()

For active deployments, the above query can provide insights. The balancing process, which originates on a randomly
selected mongos, takes a special balancer lock that prevents other balancing activity from transpiring. Use the
following command, also to the config database, to check the status of the balancer lock.
db.locks.find( { _id : "balancer" } )

If this lock exists, make sure that the balancer process is actively using this lock.
Run-time Database Configuration
The command line and configuration file interfaces provide MongoDB administrators with a large number of options and settings for controlling the operation of the database system. This document provides an overview
of common configurations and examples of best-practice configurations for common use cases.
While both interfaces provide access to the same collection of options and settings, this document primarily uses the
configuration file interface. If you run MongoDB using a control script or installed from a package for your operating
system, you likely already have a configuration file located at /etc/mongodb.conf. Confirm this by checking the
contents of the /etc/init.d/mongod or /etc/rc.d/mongod script to ensure that the control scripts start the
mongod with the appropriate configuration file (see below.)
To start a MongoDB instance using this configuration issue a command in the following form:
mongod --config /etc/mongodb.conf
mongod -f /etc/mongodb.conf

Modify the values in the /etc/mongodb.conf file on your system to control the configuration of your database
instance.
54 http://mms.mongodb.com

146

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Configure the Database

Consider the following basic configuration:

fork = true
bind_ip = 127.0.0.1
port = 27017
quiet = true
dbpath = /srv/mongodb
logpath = /var/log/mongodb/mongod.log
logappend = true
journal = true

For most standalone servers, this is a sufficient base configuration. It makes several assumptions, but consider the
following explanation:
fork is true, which enables a daemon mode for mongod, which detaches (i.e. forks) the MongoDB from
the current session and allows you to run the database as a conventional server.
bind_ip is 127.0.0.1, which forces the server to only listen for requests on the localhost IP. Only bind to
secure interfaces that the application-level systems can access with access control provided by system network
filtering (i.e. firewall).
port is 27017, which is the default MongoDB port for database instances. MongoDB can bind to any port.
You can also filter access based on port using network filtering tools.
Note: UNIX-like systems require superuser privileges to attach processes to ports lower than 1024.
quiet is true. This disables all but the most critical entries in output/log file. In normal operation this is
the preferable operation to avoid log noise. In diagnostic or testing situations, set this value to false. Use
setParameter to modify this setting during run time.
dbpath is /srv/mongodb, which specifies where MongoDB will store its data files. /srv/mongodb and
/var/lib/mongodb are popular locations. The user account that mongod runs under will need read and
write access to this directory.
logpath is /var/log/mongodb/mongod.log which is where mongod will write its output. If you do
not set this value, mongod writes all output to standard output (e.g. stdout.)
logappend is true, which ensures that mongod does not overwrite an existing log file following the server
start operation.
journal is true, which enables journaling. Journaling ensures single instance write-durability. 64-bit builds
of mongod enable journaling by default. Thus, this setting may be redundant.
Given the default configuration, some of these values may be redundant. However, in many situations explicitly stating
the configuration increases overall system intelligibility.
Security Considerations

The following collection of configuration options are useful for limiting access to a mongod instance. Consider the
following:
bind_ip = 127.0.0.1,10.8.0.10,192.168.4.24
nounixsocket = true
auth = true

Consider the following explanation for these configuration decisions:

4.1. Administration Concepts

147

MongoDB Documentation, Release 2.4.9

bind_ip has three values: 127.0.0.1, the localhost interface; 10.8.0.10, a private IP address typically
used for local networks and VPN interfaces; and 192.168.4.24, a private network interface typically used
for local networks.
Because production MongoDB instances need to be accessible from multiple database servers, it is important
to bind MongoDB to multiple interfaces that are accessible from your application servers. At the same time its
important to limit these interfaces to interfaces controlled and protected at the network layer.
nounixsocket to true disables the UNIX Socket, which is otherwise enabled by default. This limits
access on the local system. This is desirable when running MongoDB on systems with shared access, but in
most situations has minimal impact.
auth is true enables the authentication system within MongoDB. If enabled you will need to log in by
connecting over the localhost interface for the first time to create user credentials.
See also:
Security Concepts (page 239)
Replication and Sharding Configuration

Replication Configuration Replica set configuration is straightforward, and only requires that the replSet have
a value that is consistent among all members of the set. Consider the following:
replSet = set0

Use descriptive names for sets. Once configured use the mongo shell to add hosts to the replica set.
See also:
Replica set reconfiguration (page 470).
To enable authentication for the replica set, add the following option:
keyFile = /srv/mongodb/keyfile

New in version 1.8: for replica sets, and 1.9.1 for sharded replica sets.
Setting keyFile enables authentication and specifies a key file for the replica set member use to when authenticating
to each other. The content of the key file is arbitrary, but must be the same on all members of the replica set and
mongos instances that connect to the set. The keyfile must be less than one kilobyte in size and may only contain
characters in the base64 set and the file must not have group or world permissions on UNIX systems.
See also:
The Replica set Reconfiguration (page 470) section for information regarding the process for changing replica set
during operation.
Additionally, consider the Replica Set Security (page 240) section for information on configuring authentication with
replica sets.
Finally, see the Replication (page 377) document for more information on replication in MongoDB and replica set
configuration in general.
Sharding Configuration Sharding requires a number of mongod instances with different configurations. The config servers store the clusters metadata, while the cluster distributes data among one or more shard servers.
Note: Config servers are not replica sets.

148

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

To set up one or three config server instances as normal (page 147) mongod instances, and then add the following
configuration option:
configsvr = true
bind_ip = 10.8.0.12
port = 27001

This creates a config server running on the private IP address 10.8.0.12 on port 27001. Make sure that there are
no port conflicts, and that your config server is accessible from all of your mongos and mongod instances.
To set up shards, configure two or more mongod instance using your base configuration (page 147), adding the
shardsvr setting:
shardsvr = true

Finally, to establish the cluster, configure at least one mongos process with the following settings:
configdb = 10.8.0.12:27001
chunkSize = 64

You can specify multiple configdb instances by specifying hostnames and ports in the form of a comma separated
list. In general, avoid modifying the chunkSize from the default value of 64, 55 and should ensure this setting is
consistent among all mongos instances.
See also:
The Sharding (page 479) section of the manual for more information on sharding and cluster configuration.
Run Multiple Database Instances on the Same System

In many cases running multiple instances of mongod on a single system is not recommended. On some types of
deployments 56 and for testing purposes you may need to run more than one mongod on a single system.
In these cases, use a base configuration (page 147) for each instance, but consider the following configuration values:
dbpath = /srv/mongodb/db0/
pidfilepath = /srv/mongodb/db0.pid

The dbpath value controls the location of the mongod instances data directory. Ensure that each database has a
distinct and well labeled data directory. The pidfilepath controls where mongod process places its process id
file. As this tracks the specific mongod file, it is crucial that file be unique and well labeled to make it easy to start
and stop these processes.
Create additional control scripts and/or adjust your existing MongoDB configuration and control script as needed to
control these processes.
Diagnostic Configurations

The following configuration options control various mongod behaviors for diagnostic purposes. The following settings have default values that tuned for general production purposes:
55 Chunk size is 64 megabytes by default, which provides the ideal balance between the most even distribution of data, for which smaller chunk
sizes are best, and minimizing chunk migration, for which larger chunk sizes are optimal.
56 Single-tenant systems with SSD or other high performance disks may provide acceptable performance levels for multiple mongod instances.
Additionally, you may find that multiple databases with small working sets may function acceptably on a single system.

4.1. Administration Concepts

149

MongoDB Documentation, Release 2.4.9

slowms = 50
profile = 3
verbose = true
diaglog = 3
objcheck = true
cpu = true

Use the base configuration (page 147) and add these options if you are experiencing some unknown issue or performance problem as needed:
slowms configures the threshold for the database profiler to consider a query slow. The default value is
100 milliseconds. Set a lower value if the database profiler does not return useful results. See Optimization
Strategies for MongoDB (page 165) for more information on optimizing operations in MongoDB.
profile sets the database profiler level. The profiler is not active by default because of the possible impact
on the profiler itself on performance. Unless this setting has a value, queries are not profiled.
verbose enables a verbose logging mode that modifies mongod output and increases logging to include a
greater number of events. Only use this option if you are experiencing an issue that is not reflected in the normal
logging level. If you require additional verbosity, consider the following options:
v = true
vv = true
vvv = true
vvvv = true
vvvvv = true

Each additional level v adds additional verbosity to the logging. The verbose option is equal to v = true.
diaglog enables diagnostic logging. Level 3 logs all read and write options.
objcheck forces mongod to validate all requests from clients upon receipt. Use this option to ensure that
invalid requests are not causing errors, particularly when running a database with untrusted clients. This option
may affect database performance.
cpu forces mongod to report the percentage of the last interval spent in write lock. The interval is typically 4
seconds, and each output line in the log includes both the actual interval since the last report and the percentage
of time spent in write lock.
Import and Export MongoDB Data
This document provides an overview of the import and export programs included in the MongoDB distribution. These
tools are useful when you want to backup or export a portion of your data without capturing the state of the entire
database, or for simple data ingestion cases. For more complex data migration tasks, you may want to write your own
import and export scripts using a client driver to interact with the database itself. For disaster recovery protection and
routine database backup operation, use full database instance backups (page 136).
Warning: Because these tools primarily operate by interacting with a running mongod instance, they can impact
the performance of your running database.
Not only do these processes create traffic for a running database instance, they also force the database to read all
data through memory. When MongoDB reads infrequently used data, it can supplant more frequently accessed
data, causing a deterioration in performance for the databases regular workload.
mongoimport and mongoexport do not reliably preserve all rich BSON data types, because BSON is a superset of JSON. Thus, mongoimport and mongoexport cannot represent BSON data accurately in JSON. As
a result data exported or imported with these tools may lose some measure of fidelity. See MongoDB Extended
JSON (page 227) for more information about MongoDB Extended JSON.

150

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

See also:
See the MongoDB Backup Methods (page 136) document or the MMS Backup Manual57 for more information on
backing up MongoDB instances. Additionally, consider the following references for commands addressed in this
document:
http://docs.mongodb.org/manualreference/program/mongoexport
http://docs.mongodb.org/manualreference/program/mongorestore
http://docs.mongodb.org/manualreference/program/mongodump
If you want to transform and process data once youve imported it in MongoDB consider the documents in the Aggregation (page 277) section, including:
Map-Reduce (page 284) and
Aggregation Concepts (page 281).
Data Type Fidelity

JSON does not have the following data types that exist in BSON documents: data_binary, data_date,
data_timestamp, data_regex, data_oid and data_ref. As a result using any tool that decodes BSON
documents into JSON will suffer some loss of fidelity.
If maintaining type fidelity is important, consider writing a data import and export system that does not force BSON
documents into JSON form as part of the process. The following list of types contain examples for how MongoDB
will represent how BSON documents render in JSON.
data_binary
{ "$binary" : "<bindata>", "$type" : "<t>" }

<bindata> is the base64 representation of a binary string. <t> is the hexadecimal representation of a single
byte indicating the data type.
data_date
Date( <date> )

<date> is the JSON representation of a 64-bit signed integer for milliseconds since epoch.
data_timestamp
Timestamp( <t>, <i> )

<t> is the JSON representation of a 32-bit unsigned integer for milliseconds since epoch. <i> is a 32-bit
unsigned integer for the increment.
data_regex
/<jRegex>/<jOptions>

<jRegex> is a string that may contain valid JSON characters and unescaped double quote (i.e. ") characters,
but may not contain unescaped forward slash (i.e. http://docs.mongodb.org/manual) characters.
<jOptions> is a string that may contain only the characters g, i, m, and s.
data_oid
ObjectId( "<id>" )
57 https://mms.mongodb.com/help/backup

4.1. Administration Concepts

151

MongoDB Documentation, Release 2.4.9

<id> is a 24 character hexadecimal string. These representations require that data_oid values have an
associated field named _id.
data_ref
DBRef( "<name>", "<id>" )

<name> is a string of valid JSON characters. <id> is a 24 character hexadecimal string.

See also:
MongoDB Extended JSON (page 227)
Data Import and Export and Backups Operations

For resilient and non-disruptive backups, use a file system or block-level disk snapshot function, such as the methods described in the MongoDB Backup Methods (page 136) document. The tools and operations discussed provide
functionality thats useful in the context of providing some kinds of backups.
By contrast, use import and export tools to backup a small subset of your data or to move data to or from a 3rd party
system. These backups may capture a small crucial set of data or a frequently modified section of data, for extra
insurance, or for ease of access. No matter how you decide to import or export your data, consider the following
guidelines:
Label files so that you can identify what point in time the export or backup reflects.
Labeling should describe the contents of the backup, and reflect the subset of the data corpus, captured in the
backup or export.
Do not create or apply exports if the backup process itself will have an adverse effect on a production system.
Make sure that they reflect a consistent data state. Export or backup processes can impact data integrity (i.e.
type fidelity) and consistency if updates continue during the backup process.
Test backups and exports by restoring and importing to ensure that the backups are useful.
Human Intelligible Import/Export Formats

This section describes a process to import/export your database, or a portion thereof, to a file in a JSON or CSV format.
See also:
The
http://docs.mongodb.org/manualreference/program/mongoimport
and
http://docs.mongodb.org/manualreference/program/mongoexport documents contain complete
documentation of these tools. If you have questions about the function and parameters of these tools not covered here,
please refer to these documents.
If you want to simply copy a database or collection from one instance to another, consider using the copydb,
clone, or cloneCollection commands, which may be more suited to this task. The mongo shell provides
the db.copyDatabase() method.
These tools may also be useful for importing data into a MongoDB database from third party applications.
Collection Export with mongoexport With the mongoexport utility you can create a backup file. In the most
simple invocation, the command takes the following form:
mongoexport --collection collection --out collection.json

152

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

This will export all documents in the collection named collection into the file collection.json. Without
the output specification (i.e. --out collection.json), mongoexport writes output to standard output (i.e.
stdout). You can further narrow the results by supplying a query filter using the --query and limit results to a
single database using the --db option. For instance:
mongoexport --db sales --collection contacts --query '{"field": 1}'

This command returns all documents in the sales databases contacts collection, with a field named field with
a value of 1. Enclose the query in single quotes (e.g. ) to ensure that it does not interact with your shell environment.
The resulting documents will return on standard output.
By default, mongoexport returns one JSON document per MongoDB document. Specify the --jsonArray
argument to return the export as a single JSON array. Use the --csv file to return the result in CSV (comma
separated values) format.
If your mongod instance is not running, you can use the --dbpath option to specify the location to your MongoDB instances database files. See the following example:
mongoexport --db sales --collection contacts --dbpath /srv/MongoDB/

This reads the data files directly. This locks the data directory to prevent conflicting writes. The mongod process must
not be running or attached to these data files when you run mongoexport in this configuration.
The --host and --port options allow you to specify a non-local host to connect to capture the export. Consider
the following example:

mongoexport --host mongodb1.example.net --port 37017 --username user --password pass --collection con

On any mongoexport command you may, as above specify username and password credentials as above.
Collection Import with mongoimport To restore a backup taken with mongoexport. Most of the arguments
to mongoexport also exist for mongoimport. Consider the following command:
mongoimport --collection collection --file collection.json

This imports the contents of the file collection.json into the collection named collection. If you do not
specify a file with the --file option, mongoimport accepts input over standard input (e.g. stdin.)
If you specify the --upsert option, all of mongoimport operations will attempt to update existing documents
in the database and insert other documents. This option will cause some performance impact depending on your
configuration.
You can specify the database option --db to import these documents to a particular database. If your MongoDB
instance is not running, use the --dbpath option to specify the location of your MongoDB instances database
files. Consider using the --journal option to ensure that mongoimport records its operations in the journal. The mongod process must not be running or attached to these data files when you run mongoimport in this
configuration.
Use the --ignoreBlanks option to ignore blank fields. For CSV and TSV imports, this option provides the
desired functionality in most cases: it avoids inserting blank fields in MongoDB documents.
Production Notes
This page details system configurations that affect MongoDB, especially in production.
Note: MongoDB Management Service (MMS)58 is a hosted monitoring service which collects and aggregates diag-

4.1. Administration Concepts

153

MongoDB Documentation, Release 2.4.9

nostic data to provide insight into the performance and operation of MongoDB deployments. See the MMS Website59
and the MMS documentation60 for more information.

Packages

MongoDB Be sure you have the latest stable release. All releases are available on the Downloads61 page. This is a
good place to verify what is current, even if you then choose to install via a package manager.
Always use 64-bit builds for production. The 32-bit build MongoDB offers for test and development environments is
not suitable for production deployments as it can store no more than 2GB of data. See the 32-bit limitations (page 558)
for more information.
32-bit builds exist to support use on development machines.
Operating Systems MongoDB distributions are currently available for Mac OS X, Linux, Windows Server 2008 R2
64bit, Windows 7 (32 bit and 64 bit), Windows Vista, and Solaris platforms.
Note: MongoDB uses the GNU C Library62 (glibc) if available on a system. MongoDB requires version at least
glibc-2.12-1.2.el6 to avoid a known bug with earlier versions. For best results use at least version 2.13.

Concurrency

In earlier versions of MongoDB, all write operations contended for a single readers-writer lock on the MongoDB
instance. As of version 2.2, each database has a readers-writer lock that allows concurrent reads access to a database,
but gives exclusive access to a single write operation per database. See the Concurrency (page 569) page for more
information.
Journaling

MongoDB uses write ahead logging to an on-disk journal to guarantee that MongoDB is able to quickly recover the
write operations (page 42) following a crash or other serious failure.
In order to ensure that mongod will be able to recover and remain in a valid state following a crash, you should leave
journaling enabled. See Journaling (page 234) for more information.
Networking

Use Trusted Networking Environments Always run MongoDB in a trusted environment, with network rules that
prevent access from all unknown machines, systems, and networks. As with any sensitive system dependent on
network access, your MongoDB deployment should only be accessible to specific systems that require access, such as
application servers, monitoring services, and other MongoDB components.
Note: By default, auth is not enabled and mongod assumes a trusted environment. You can enable security/auth
(page 239) mode if you need it.
58 http://mms.mongodb.com
59 http://mms.mongodb.com/
60 http://mms.mongodb.com/help/
61 http://www.mongodb.org/downloads
62 http://www.gnu.org/software/libc/

154

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

See documents in the Security (page 237) section for additional information, specifically:
Configuration Options (page 242)
Firewalls (page 243)
Configure Linux iptables Firewall for MongoDB (page 245)
Configure Windows netsh Firewall for MongoDB (page 249)
For Windows users, consider the Windows Server Technet Article on TCP Configuration63 when deploying MongoDB
on Windows.
Connection Pools To avoid overloading the connection resources of a single mongod or mongos instance, ensure
that clients maintain reasonable connection pool sizes.
The connPoolStats database command returns information regarding the number of open connections to the
current database for mongos instances and mongod instances in sharded clusters.
Hardware Considerations

MongoDB is designed specifically with commodity hardware in mind and has few hardware requirements or limitations. MongoDBs core components run on little-endian hardware, primarily x86/x86_64 processors. Client libraries
(i.e. drivers) can run on big or little endian systems.
Hardware Requirements and Limitations
following properties:
Allocate Sufficient RAM and CPU
for performance.

The hardware for the most effective MongoDB deployments have the

As with all software, more RAM and a faster CPU clock speed are important

In general, databases are not CPU bound. As such, increasing the number of cores can help, but does not provide
significant marginal return.
Use Solid State Disks (SSDs)
(Solid State Disk).

MongoDB has good results and a good price-performance ratio with SATA SSD

Use SSD if available and economical. Spinning disks can be performant, but SSDs capacity for random I/O operations
works well with the update model of mongod.
Commodity (SATA) spinning drives are often a good option, as the increase to random I/O for more expensive drives
is not that dramatic (only on the order of 2x). Using SSDs or increasing RAM may be more effective in increasing I/O
throughput.
Avoid Remote File Systems
Remote file storage can create performance problems in MongoDB. See Remote Filesystems (page 156) for
more information about storage and MongoDB.
63 http://technet.microsoft.com/en-us/library/dd349797.aspx

4.1. Administration Concepts

155

MongoDB Documentation, Release 2.4.9

MongoDB and NUMA Hardware

Important: The discussion of NUMA in this section only applies to Linux, and therefore does not affect deployments
where mongod instances run other UNIX-like systems or on Windows.
Running MongoDB on a system with Non-Uniform Access Memory (NUMA) can cause a number of operational
problems, including slow performance for periods of time or high system process usage.
When running MongoDB on NUMA hardware, you should disable NUMA for MongoDB and instead set an interleave
memory policy.
Note: MongoDB version 2.0 and greater checks these settings on start up when deployed on a Linux-based system,
and prints a warning if the system is NUMA-based.
To disable NUMA for MongoDB and set an interleave memory policy, use the numactl command and start mongod
in the following manner:
numactl --interleave=all /usr/bin/local/mongod

Then, disable zone reclaim in the proc settings using the following command:
echo 0 > /proc/sys/vm/zone_reclaim_mode

To fully disable NUMA, you must perform both operations. For more information, see the Documentation for
/proc/sys/vm/*64 .
See the The MySQL swap insanity problem and the effects of NUMA65 post, which describes the effects of NUMA
on databases. This blog post addresses the impact of NUMA for MySQL, but the issues for MongoDB are similar. The
post introduces NUMA and its goals, and illustrates how these goals are not compatible with production databases.
Disk and Storage Systems
Swap Assign swap space for your systems. Allocating swap space can avoid issues with memory contention and
can prevent the OOM Killer on Linux systems from killing mongod.
The method mongod uses to map memory files to memory ensures that the operating system will never store MongoDB data in swap space.
RAID Most MongoDB deployments should use disks backed by RAID-10.
RAID-5 and RAID-6 do not typically provide sufficient performance to support a MongoDB deployment.
Avoid RAID-0 with MongoDB deployments. While RAID-0 provides good write performance, it also provides limited
availability and can lead to reduced performance on read operations, particularly when using Amazons EBS volumes.
Remote Filesystems The Network File System protocol (NFS) is not recommended for use with MongoDB as some
versions perform poorly.
Performance problems arise when both the data files and the journal files are hosted on NFS. You may experience
better performance if you place the journal on local or iscsi volumes. If you must use NFS, add the following NFS
options to your /etc/fstab file: bg, nolock, and noatime.
64 http://www.kernel.org/doc/Documentation/sysctl/vm.txt
65 http://jcole.us/blog/archives/2010/09/28/mysql-swap-insanity-and-the-numa-architecture/

156

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Separate Components onto Different Storage Devices For improved performance, consider separating your
databases data, journal, and logs onto different storage devices, based on your applications access and write pattern.
Note: This will affect your ability to create snapshot-style backups of your data, since the files will be on different
devices and volumes.

Architecture

Write Concern Write concern describes the guarantee that MongoDB provides when reporting on the success of
a write operation. The strength of the write concerns determine the level of guarantee. When inserts, updates and
deletes have a weak write concern, write operations return quickly. In some failure cases, write operations issued with
weak write concerns may not persist. With stronger write concerns, clients wait after sending a write operation for
MongoDB to confirm the write operations.
MongoDB provides different levels of write concern to better address the specific needs of applications. Clients
may adjust write concern to ensure that the most important operations persist successfully to an entire MongoDB
deployment. For other less critical operations, clients can adjust the write concern to ensure faster performance rather
than ensure persistence to the entire deployment.
See Write Concern (page 47) for more information about choosing an appropriate write concern level for your deployment.
Replica Sets See Replica Set Deployment Architectures (page 390) for an overview of architectural considerations
for replica set deployments.
Sharded Clusters See Production Cluster Architecture (page 490) for an overview of recommended sharded cluster
architectures for production deployments.
Platforms

MongoDB on Linux
Important: The following discussion only applies to Linux, and therefore does not affect deployments where
mongod instances run other UNIX-like systems or on Windows.

Kernel and File Systems When running MongoDB in production on Linux, it is recommended that you use Linux
kernel version 2.6.36 or later.
MongoDB preallocates its database files before using them and often creates large files. As such, you should use the
Ext4 and XFS file systems:
In general, if you use the Ext4 file system, use at least version 2.6.23 of the Linux Kernel.
In general, if you use the XFS file system, use at least version 2.6.25 of the Linux Kernel.
Some Linux distributions require different versions of the kernel to support using ext4 and/or xfs:

4.1. Administration Concepts

157

MongoDB Documentation, Release 2.4.9

Linux Distribution
CentOS 5.5
CentOS 5.6
CentOS 5.8
CentOS 6.1
RHEL 5.6
RHEL 6.0
Ubuntu 10.04.4 LTS
Amazon Linux AMI release 2012.03

Filesystem
ext4, xfs
ext4, xfs
ext4, xfs
ext4, xfs
ext4
xfs
ext4, xfs
ext4

Kernel Version
2.6.18-194.el5
2.6.18-238.el5
2.6.18-308.8.2.el5
2.6.32-131.0.15.el6.x86_64
2.6.18-238
2.6.32-71
2.6.32-38-server
3.2.12-3.2.4.amzn1.x86_64

Important: MongoDB requires a filesystem that supports fsync() on directories. For example, HGFS and Virtual
Boxs shared folders do not support this operation.

Recommended Configuration
Turn off atime for the storage volume containing the database files.
Set the file descriptor limit, -n, and the user process limit (ulimit), -u, above 20,000, according to the suggestions in the UNIX ulimit Settings (page 223). A low ulimit will affect MongoDB when under heavy use and can
produce errors and lead to failed connections to MongoDB processes and loss of service.
Disable transparent huge pages as MongoDB performs better with normal (4096 bytes) virtual memory pages.
Disable NUMA in your BIOS. If that is not possible see MongoDB on NUMA Hardware (page 156).
Ensure that readahead settings for the block devices that store the database files are appropriate. For random
access use patterns, set low readahead values. A readahead of 32 (16kb) often works well.
Use the Network Time Protocol (NTP) to synchronize time among your hosts. This is especially important in
sharded clusters.
MongoDB on Virtual Environments
more common virtual environments.

The section describes considerations when running MongoDB in some of the

EC2 MongoDB is compatible with EC2 and requires no configuration changes specific to the environment.
You may alternately choose to obtain a set of Amazon Machine Images (AMI) that bundle together MongoDB and
Amazons Provisioned IOPS storage volumes. Provisioned IOPS can greatly increase MongoDBs performance and
ease of use. For more information, see this blog post66 .
VMWare MongoDB is compatible with VMWare. As some users have run into issues with VMWares memory
overcommit feature, disabling the feature is recommended.
It is possible to clone a virtual machine running MongoDB. You might use this function to spin up a new virtual host
to add as a member of a replica set. If you clone a VM with journaling enabled, the clone snapshot will be valid. If
not using journaling, first stop mongod, then clone the VM, and finally, restart mongod.
OpenVZ Some users have had issues when running MongoDB on some older version of OpenVZ due to its handling
of virtual memory, as with VMWare.
This issue seems to have been resolved in the more recent versions of OpenVZ.
66 http://www.mongodb.com/blog/post/provisioned-iops-aws-marketplace-significantly-boosts-mongodb-performance-ease-use

158

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Performance Monitoring

iostat On Linux, use the iostat command to check if disk I/O is a bottleneck for your database. Specify a number
of seconds when running iostat to avoid displaying stats covering the time since server boot.
For example, the following command will display extended statistics and the time for each displayed report, with
traffic in MB/s, at one second intervals:
iostat -xmt 1

Key fields from iostat:

%util: this is the most useful field for a quick check, it indicates what percent of the time the device/drive is
in use.
avgrq-sz: average request size. Smaller number for this value reflect more random IO operations.
bwm-ng bwm-ng67 is a command-line tool for monitoring network use. If you suspect a network-based bottleneck,
you may use bwm-ng to begin your diagnostic process.
Backups

To make backups of your MongoDB database, please refer to MongoDB Backup Methods (page 136).

4.1.2 Data Management

These document introduce data management practices and strategies for MongoDB deployments, including strategies
for managing multi-data center deployments, managing larger file stores, and data lifecycle tools.
Data Center Awareness (page 159) Presents the MongoDB features that allow application developers and database
administrators to configure their deployments to be more data center aware or allow operational and locationbased separation.
Capped Collections (page 161) Capped collections provide a special type of size-constrained collections that preserve
insertion order and can support high volume inserts.
Expire Data from Collections by Setting TTL (page 163) TTL collections make it possible to automatically remove
data from a collection based on the value of a timestamp and are useful for managing data like machine generated
event data that are only useful for a limited period of time.
Data Center Awareness
MongoDB provides a number of features that allow application developers and database administrators to customize
the behavior of a sharded cluster or replica set deployment so that MongoDB may be more data center aware, or
allow operational and location-based separation.
MongoDB also supports segregation based on functional parameters, to ensure that certain mongod instances are
only used for reporting workloads or that certain high-frequency portions of a sharded collection only exist on specific
shards.
The following documents, found either in this section or other sections of this manual, provide information on customizing a deployment for operation- and location-based separation:
67 http://www.gropp.org/?id=projects&sub=bwm-ng

4.1. Administration Concepts

159

MongoDB Documentation, Release 2.4.9

Operational Segregation in MongoDB Deployments (page 160) MongoDB lets you specify that certain application
operations use certain mongod instances.
Tag Aware Sharding (page 540) Tags associate specific ranges of shard key values with specific shards for use in
managing deployment patterns.
Manage Shard Tags (page 541) Use tags to associate specific ranges of shard key values with specific shards.
Operational Segregation in MongoDB Deployments

Operational Overview MongoDB includes a number of features that allow database administrators and developers
to segregate application operations to MongoDB deployments by functional or geographical groupings.
This capability provides data center awareness, which allows applications to target MongoDB deployments with
consideration of the physical location of the mongod instances. MongoDB supports segmentation of operations
across different dimensions, which may include multiple data centers and geographical regions in multi-data center
deployments, racks, networks, or power circuits in single data center deployments.
MongoDB also supports segregation of database operations based on functional or operational parameters, to ensure
that certain mongod instances are only used for reporting workloads or that certain high-frequency portions of a
sharded collection only exist on specific shards.
Specifically, with MongoDB, you can:
ensure write operations propagate to specific members of a replica set, or to specific members of replica sets.
ensure that specific members of a replica set respond to queries.
ensure that specific ranges of your shard key balance onto and reside on specific shards.
combine the above features in a single distributed deployment, on a per-operation (for read and write operations)
and collection (for chunk distribution in sharded clusters distribution) basis.
For full documentation of these features, see the following documentation in the MongoDB Manual:
Read Preferences (page 406), which controls how drivers help applications target read operations to members
of a replica set.
Write Concerns (page 47), which controls how MongoDB ensures that write operations propagate to members
of a replica set.
Replica Set Tags (page 450), which control how applications create and interact with custom groupings of replica
set members to create custom application-specific read preferences and write concerns.
Tag Aware Sharding (page 540), which allows MongoDB administrators to define an application-specific balancing policy, to control how documents belonging to specific ranges of a shard key distribute to shards in the
sharded cluster.
See also:
Before adding operational segregation features to your application and MongoDB deployment, become familiar with
all documentation of replication (page 377), :and doc:sharding </sharding>.
Further Reading

The Write Concern (page 47) and Read Preference (page 406) documents, which address capabilities related to
data center awareness.
Deploy a Geographically Redundant Replica Set (page 425).

160

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Capped Collections
Capped collections are fixed-size collections that support high-throughput operations that insert, retrieve, and delete
documents based on insertion order. Capped collections work in a way similar to circular buffers: once a collection
fills its allocated space, it makes room for new documents by overwriting the oldest documents in the collection.
See createCollection() or createCollection for more information on creating capped collections.
Capped collections have the following behaviors:
Capped collections guarantee preservation of the insertion order. As a result, queries do not need an index to
return documents in insertion order. Without this indexing overhead, they can support higher insertion throughput.
Capped collections guarantee that insertion order is identical to the order on disk (natural order) and do so
by prohibiting updates that increase document size. Capped collections only allow updates that fit the original
document size, which ensures a document does not change its location on disk.
Capped collections automatically remove the oldest documents in the collection without requiring scripts or
explicit remove operations.
For example, the oplog.rs collection that stores a log of the operations in a replica set uses a capped collection.
Consider the following potential use cases for capped collections:
Store log information generated by high-volume systems. Inserting documents in a capped collection without
an index is close to the speed of writing log information directly to a file system. Furthermore, the built-in
first-in-first-out property maintains the order of events, while managing storage use.
Cache small amounts of data in a capped collections. Since caches are read rather than write heavy, you would
either need to ensure that this collection always remains in the working set (i.e. in RAM) or accept some write
penalty for the required index or indexes.
Recommendations and Restrictions

You can update documents in a collection after inserting them. However, these updates cannot cause the documents to grow. If the update operation causes the document to grow beyond their original size, the update
operation will fail.
If you plan to update documents in a capped collection, create an index so that these update operations do not
require a table scan.
You cannot delete documents from a capped collection. To remove all records from a capped collection, use the
emptycapped command. To remove the collection entirely, use the drop() method.
You cannot shard a capped collection.
Capped collections created after 2.2 have an _id field and an index on the _id field by default. Capped
collections created before 2.2 do not have an index on the _id field by default. If you are using capped
collections with replication prior to 2.2, you should explicitly create an index on the _id field.
Warning: If you have a capped collection in a replica set outside of the local database, before 2.2,
you should create a unique index on _id. Ensure uniqueness using the unique: true option to
the ensureIndex() method or by using an ObjectId for the _id field. Alternately, you can use the
autoIndexId option to create when creating the capped collection, as in the Query a Capped Collection (page 162) procedure.
Use natural ordering to retrieve the most recently inserted elements from the collection efficiently. This is
(somewhat) analogous to tail on a log file.

4.1. Administration Concepts

161

MongoDB Documentation, Release 2.4.9

Procedures

Create a Capped Collection You must create capped collections explicitly using the createCollection()
method, which is a helper in the mongo shell for the create command. When creating a capped collection you must
specify the maximum size of the collection in bytes, which MongoDB will pre-allocate for the collection. The size of
the capped collection includes a small amount of space for internal overhead.
db.createCollection( "log", { capped: true, size: 100000 } )

Additionally, you may also specify a maximum number of documents for the collection using the max field as in the
following document:
db.createCollection("log", { capped : true, size : 5242880, max : 5000 } )

Important: The size argument is always required, even when you specify max number of documents. MongoDB
will remove older documents if a collection reaches the maximum size limit before it reaches the maximum document
count.
See
createCollection() and create.

Query a Capped Collection If you perform a find() on a capped collection with no ordering specified, MongoDB
guarantees that the ordering of results is the same as the insertion order.
To retrieve documents in reverse insertion order, issue find() along with the sort() method with the $natural
parameter set to -1, as shown in the following example:
db.cappedCollection.find().sort( { $natural: -1 } )

Check if a Collection is Capped Use the isCapped() method to determine if a collection is capped, as follows:
db.collection.isCapped()

Convert a Collection to Capped You can convert a non-capped collection to a capped collection with the
convertToCapped command:
db.runCommand({"convertToCapped": "mycoll", size: 100000});

The size parameter specifies the size of the capped collection in bytes.
Warning: This command obtains a global write lock and will block other operations until it has completed.
Changed in version 2.2: Before 2.2, capped collections did not have an index on _id unless you specified
autoIndexId to the create, after 2.2 this became the default.
Automatically Remove Data After a Specified Period of Time For additional flexibility when expiring data, consider MongoDBs TTL indexes, as described in Expire Data from Collections by Setting TTL (page 163). These indexes
allow you to expire and remove data from normal collections using a special type, based on the value of a date-typed
field and a TTL value for the index.
TTL Collections (page 163) are not compatible with capped collections.

162

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Tailable Cursor You can use a tailable cursor with capped collections. Similar to the Unix tail -f command,
the tailable cursor tails the end of a capped collection. As new documents are inserted into the capped collection,
you can use the tailable cursor to continue retrieving documents.
See Create Tailable Cursor (page 74) for information on creating a tailable cursor.
Expire Data from Collections by Setting TTL
New in version 2.2.
This document provides an introduction to MongoDBs time to live or TTL collection feature. TTL collections
make it possible to store data in MongoDB and have the mongod automatically remove data after a specified number
of seconds or at a specific clock time.
Data expiration is useful for some classes of information, including machine generated event data, logs, and session
information that only need to persist for a limited period of time.
A special index type supports the implementation of TTL collections. TTL relies on a background thread in mongod
that reads the date-typed values in the index and removes expired documents from the collection.
Considerations

The _id field does not support TTL indexes.

You cannot create a TTL index on a field that already has an index.
A document will not expire if the indexed field does not exist.
A document will not expire if the indexed field is not a date BSON type or an array of date BSON types.
The TTL index may not be compound (may not have multiple fields).
If the TTL field holds an array, and there are multiple date-typed data in the index, the document will expire
when the lowest (i.e. earliest) date matches the expiration threshold.
You cannot create a TTL index on a capped collection, because MongoDB cannot remove documents from a
capped collection.
You cannot use ensureIndex() to change the value of expireAfterSeconds.
collMod database command in conjunction with the index collection flag.

Instead use the

When you build a TTL index in the background (page 336), the TTL thread can begin deleting documents
while the index is building. If you build a TTL index in the foreground, MongoDB begins removing expired
documents as soon as the index finishes building.
When the TTL thread is active, you will see delete (page 42) operations in the output of db.currentOp() or in the
data collected by the database profiler (page 174).
When using TTL indexes on replica sets, the TTL background thread only deletes documents on primary members.
However, the TTL background thread does run on secondaries. Secondary members replicate deletion operations from
the primary.
The TTL index does not guarantee that expired data will be deleted immediately. There may be a delay between the
time a document expires and the time that MongoDB removes the document from the database.
The background task that removes expired documents runs every 60 seconds. As a result, documents may remain in a
collection after they expire but before the background task runs or completes.
The duration of the removal operation depends on the workload of your mongod instance. Therefore, expired data
may exist for some time beyond the 60 second period between runs of the background task.

4.1. Administration Concepts

163

MongoDB Documentation, Release 2.4.9

All collections with an index using the expireAfterSeconds option have usePowerOf2Sizes enabled. Users
cannot modify this setting. As a result of enabling usePowerOf2Sizes, MongoDB must allocate more disk space
relative to data size. This approach helps mitigate the possibility of storage fragmentation caused by frequent delete
operations and leads to more predictable storage use patterns.
Procedures

To enable TTL for a collection, use the ensureIndex() method to create a TTL index, as shown in the examples
below.
With the exception of the background thread, a TTL index supports queries in the same way normal indexes do. You
can use TTL indexes to expire documents in one of two ways, either:
remove documents a certain number of seconds after creation. The index will support queries for the creation
time of the documents. Alternately,
specify an explicit expiration time. The index will support queries for the expiration-time of the document.
Expire Documents after a Certain Number of Seconds To expire data after a certain number of seconds, create
a TTL index on a field that holds values of BSON date type or an array of BSON date-typed objects and specify a
positive non-zero value in the expireAfterSeconds field. A document will expire when the number of seconds
in the expireAfterSeconds field has passed since the time specified in its indexed field. 68
For example, the following operation creates an index on the log.events collections createdAt field and specifies the expireAfterSeconds value of 3600 to set the expiration time to be one hour after the time specified by
createdAt.
db.log.events.ensureIndex( { "createdAt": 1 }, { expireAfterSeconds: 3600 } )

When adding documents to the log.events collection, set the createdAt field to the current time:
db.log.events.insert( {
"createdAt": new Date(),
"logEvent": 2,
"logMessage": "Success!"
} )

MongoDB will automatically delete documents from the log.events collection when the documents createdAt
value 1 is older than the number of seconds specified in expireAfterSeconds.
Expire Documents at a Certain Clock Time To expire documents at a certain clock time, begin by creating a
TTL index on a field that holds values of BSON date type or an array of BSON date-typed objects and specify an
expireAfterSeconds value of 0. For each document in the collection, set the indexed date field to a value
corresponding to the time the document should expire. If the indexed date field contains a date in the past, MongoDB
considers the document expired.
For example, the following operation creates an index on the app.events collections expireAt field and specifies
the expireAfterSeconds value of 0:
db.app.events.ensureIndex( { "expireAt": 1 }, { expireAfterSeconds: 0 } )

For each document, set the value of expireAt to correspond to the time the document should expire. For instance,
the following insert() operation adds a document that should expire at July 22, 2013 14:00:00.
68 If the field contains an array of BSON date-typed objects, data expires if at least one of BSON date-typed object is older than the number of
seconds specified in expireAfterSeconds.

164

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

db.app.events.insert( {
"expireAt": new Date('July 22, 2013 14:00:00'),
"logEvent": 2,
"logMessage": "Success!"
} )

MongoDB will automatically delete documents from the app.events collection when the documents expireAt
value is older than the number of seconds specified in expireAfterSeconds, i.e. 0 seconds older in this case. As
such, the data expires at the specified expireAt value.

4.1.3 Optimization Strategies for MongoDB

There are many factors that can affect database performance and responsiveness including index use, query structure,
data models and application design, as well as operational factors such as architecture and system configuration.
This section describes techniques for optimizing application performance with MongoDB.
Evaluate Performance of Current Operations (page 165) MongoDB provides introspection tools that describe the
query execution process, to allow users to test queries and build more efficient queries.
Use Capped Collections for Fast Writes and Reads (page 166) Outlines a use case for Capped Collections
(page 161) to optimize certain data ingestion work flows.
Optimize Query Performance (page 166) Introduces the use of projections (page 33) to reduce the amount of data
MongoDB must set to clients.
Design Notes (page 167) A collection of notes related to the architecture, design, and administration of MongoDBbased applications.
Evaluate Performance of Current Operations
The following sections describe techniques for evaluating operational performance.
Use the Database Profiler to Evaluate Operations Against the Database

MongoDB provides a database profiler that shows performance characteristics of each operation against the database.
Use the profiler to locate any queries or write operations that are running slow. You can use this information, for
example, to determine what indexes to create.
For more information, see Database Profiling (page 144).
Use db.currentOp() to Evaluate mongod Operations

The db.currentOp() method reports on current operations running on a mongod instance.

Use $explain to Evaluate Query Performance

The explain() method returns statistics on a query, and reports the index MongoDB selected to fulfill the query, as
well as information about the internal operation of the query.
Example
To use explain() on a query for documents matching the expression { a:
records, use an operation that resembles the following in the mongo shell:
4.1. Administration Concepts

1 }, in the collection named

165

MongoDB Documentation, Release 2.4.9

db.records.find( { a: 1 } ).explain()

Use Capped Collections for Fast Writes and Reads

Use Capped Collections for Fast Writes

Capped Collections (page 161) are circular, fixed-size collections that keep documents well-ordered, even without the
use of an index. This means that capped collections can receive very high-speed writes and sequential reads.
These collections are particularly useful for keeping log files but are not limited to that purpose. Use capped collections
where appropriate.
Use Natural Order for Fast Reads

To return documents in the order they exist on disk, return sorted operations using the $natural operator. On a
capped collection, this also returns the documents in the order in which they were written.
Natural order does not use indexes but can be fast for operations when you want to select the first or last items on disk.
See also:
sort() and limit().
Optimize Query Performance
Create Indexes to Support Queries

For commonly issued queries, create indexes (page 313). If a query searches multiple fields, create a compound index
(page 322). Scanning an index is much faster than scanning a collection. The indexes structures are smaller than the
documents reference, and store references in order.
Example
If you have a posts collection containing blog posts, and if you regularly issue a query that sorts on the
author_name field, then you can optimize the query by creating an index on the author_name field:
db.posts.ensureIndex( { author_name : 1 } )

Indexes also improve efficiency on queries that routinely sort on a given field.
Example
If you regularly issue a query that sorts on the timestamp field, then you can optimize the query by creating an
index on the timestamp field:
Creating this index:
db.posts.ensureIndex( { timestamp : 1 } )

Optimizes this query:

db.posts.find().sort( { timestamp : -1 } )

166

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Because MongoDB can read indexes in both ascending and descending order, the direction of a single-key index does
not matter.
Indexes support queries, update operations, and some phases of the aggregation pipeline (page 283).
Index keys that are of the BinData type are more efficiently stored in the index if:
the binary subtype value is in the range of 0-7 or 128-135, and
the length of the byte array is: 0, 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 16, 20, 24, or 32.
Limit the Number of Query Results to Reduce Network Demand

MongoDB cursors return results in groups of multiple documents. If you know the number of results you want, you
can reduce the demand on network resources by issuing the limit() method.
This is typically used in conjunction with sort operations. For example, if you need only 10 results from your query to
the posts collection, you would issue the following command:
db.posts.find().sort( { timestamp : -1 } ).limit(10)

For more information on limiting results, see limit()

Use Projections to Return Only Necessary Data

When you need only a subset of fields from documents, you can achieve better performance by returning only the
fields you need:
For example, if in your query to the posts collection, you need only the timestamp, title, author, and
abstract fields, you would issue the following command:
db.posts.find( {}, { timestamp : 1 , title : 1 , author : 1 , abstract : 1} ).sort( { timestamp : -1

For more information on using projections, see Limit Fields to Return from a Query (page 64).
Use $hint to Select a Particular Index

In most cases the query optimizer (page 37) selects the optimal index for a specific operation; however, you can force
MongoDB to use a specific index using the hint() method. Use hint() to support performance testing, or on
some queries where you must select a field or field included in several indexes.
Use the Increment Operator to Perform Operations Server-Side

Use MongoDBs $inc operator to increment or decrement values in documents. The operator increments the value
of the field on the server side, as an alternative to selecting a document, making simple modifications in the client
and then writing the entire document to the server. The $inc operator can also help avoid race conditions, which
would result when two application instances queried for a document, manually incremented a field, and saved the
entire document back at the same time.
Design Notes
This page details features of MongoDB that may be important to bear in mind when designing your applications.

4.1. Administration Concepts

167

MongoDB Documentation, Release 2.4.9

Schema Considerations

Dynamic Schema Data in MongoDB has a dynamic schema. Collections do not enforce document structure. This
facilitates iterative development and polymorphism. Nevertheless, collections often hold documents with highly homogeneous structures. See Data Modeling Concepts (page 99) for more information.
Some operational considerations include:
the exact set of collections to be used;
the indexes to be used: with the exception of the _id index, all indexes must be created explicitly;
shard key declarations: choosing a good shard key is very important as the shard key cannot be changed once
set.
Avoid importing unmodified data directly from a relational database. In general, you will want to roll up certain
data into richer documents that take advantage of MongoDBs support for sub-documents and nested arrays.
Case Sensitive Strings MongoDB strings are case sensitive. So a search for "joe" will not find "Joe".
Consider:
storing data in a normalized case format, or
using regular expressions ending with http://docs.mongodb.org/manuali, and/or
using $toLower or $toUpper in the aggregation framework (page 281).
Type Sensitive Fields MongoDB data is stored in the BSON69 format, a binary encoded serialization of JSON-like
documents. BSON encodes additional type information. See bsonspec.org70 for more information.
Consider the following document which has a field x with the string value "123":
{ x : "123" }

Then the following query which looks for a number value 123 will not return that document:
db.mycollection.find( { x : 123 } )

General Considerations

By Default, Updates Affect one Document To update multiple documents that meet your query criteria, set the
update multi option to true or 1. See: Update Multiple Documents (page 45).
Prior to MongoDB 2.2, you would specify the upsert and multi options in the update method as positional
boolean options. See: the update method reference documentation.
BSON Document Size Limit The BSON Document Size limit is currently set at 16MB per document. If you
require larger documents, use GridFS (page 104).
No Fully Generalized Transactions MongoDB does not have fully generalized transactions (page 76). If you
model your data using rich documents that closely resemble your applications objects, each logical object will be in
one MongoDB document. MongoDB allows you to modify a document in a single atomic operation. These kinds of
data modification pattern covers most common uses of transactions in other systems.
69 http://docs.mongodb.org/meta-driver/latest/legacy/bson/
70 http://bsonspec.org/#/specification

168

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Replica Set Considerations

Use an Odd Number of Replica Set Members Replica sets (page 377) perform consensus elections. To ensure
that elections will proceed successfully, either use an odd number of members, typically three, or else use an arbiter
to ensure an odd number of votes.
Keep Replica Set Members Up-to-Date MongoDB replica sets support automatic failover (page 397). It is important for your secondaries to be up-to-date. There are various strategies for assessing consistency:
1. Use monitoring tools to alert you to lag events. See Monitoring for MongoDB (page 138) for a detailed discussion of MongoDBs monitoring options.
2. Specify appropriate write concern.
3. If your application requires manual fail over, you can configure your secondaries as priority 0 (page 386).
Priority 0 secondaries require manual action for a failover. This may be practical for a small replica set, but
large deployments should fail over automatically.
See also:
replica set rollbacks (page 401).
Sharding Considerations

Pick your shard keys carefully. You cannot choose a new shard key for a collection that is already sharded.
Shard key values are immutable.
When enabling sharding on an existing collection, MongoDB imposes a maximum size on those collections to ensure that it is possible to create chunks. For a detailed explanation of this limit, see:
<sharding-existing-collection-data-size>.
To shard large amounts of data, create a new empty sharded collection, and ingest the data from the source
collection using an application level import operation.
Unique indexes are not enforced across shards except for the shard key itself. See Enforce Unique Keys for
Sharded Collections (page 542).
Consider pre-splitting (page 506) a sharded collection before a massive bulk import.

4.2 Administration Tutorials

The administration tutorials provide specific step-by-step instructions for performing common MongoDB setup, maintenance, and configuration operations.
Configuration, Maintenance, and Analysis (page 170) Describes routine management operations, including configuration and performance analysis.
Manage mongod Processes (page 171) Start, configure, and manage running mongod process.
Rotate Log Files (page 180) Archive the current log files and start new ones.
Backup and Recovery (page 190) Outlines procedures for data backup and restoration with mongod instances and
deployments.
Backup and Restore with Filesystem Snapshots (page 190) An outline of procedures for creating MongoDB
data set backups using system-level file snapshot tool, such as LVM or native storage appliance tools.

4.2. Administration Tutorials

169

MongoDB Documentation, Release 2.4.9

Backup and Restore Sharded Clusters (page 198) Detailed procedures and considerations for backing up
sharded clusters and single shards.
Recover Data after an Unexpected Shutdown (page 203) Recover data from MongoDB data files that were not
properly closed or have an invalid state.
MongoDB Scripting (page 205) An introduction to the scripting capabilities of the mongo shell and the scripting
capabilities embedded in MongoDB instances.
MongoDB Tutorials (page 186) A complete list of tutorials in the MongoDB Manual that address MongoDB operation and use.

4.2.1 Configuration, Maintenance, and Analysis

The following tutorials describe routine management operations, including configuration and performance analysis:
Use Database Commands (page 170) The process for running database commands that provide basic database operations.
Manage mongod Processes (page 171) Start, configure, and manage running mongod process.
Analyze Performance of Database Operations (page 174) Collect data that introspects the performance of query and
update operations on a mongod instance.
Monitor MongoDB with SNMP (page 177) The SNMP extension, available in MongoDB Enterprise, allows MongoDB to report data into SNMP traps.
Rotate Log Files (page 180) Archive the current log files and start new ones.
Manage Journaling (page 181) Describes the procedures for configuring and managing MongoDBs journaling system which allows MongoDB to provide crash resiliency and durability.
Store a JavaScript Function on the Server (page 183) Describes how to store JavaScript functions on a MongoDB
server.
Upgrade to the Latest Revision of MongoDB (page 184) Introduces the basic process for upgrading a MongoDB deployment between different minor release versions.
MongoDB Tutorials (page 186) A complete list of tutorials in the MongoDB Manual that address MongoDB operation and use.
Use Database Commands
The MongoDB command interface provides access to all non CRUD database operations. Fetching server stats,
initializing a replica set, and running a map-reduce job are all accomplished with commands.
See http://docs.mongodb.org/manualreference/command for list of all commands sorted by function,
and http://docs.mongodb.org/manualreference/command for a list of all commands sorted alphabetically.
Database Command Form

You specify a command first by constructing a standard BSON document whose first key is the name of the command.
For example, specify the isMaster command using the following BSON document:
{ isMaster: 1 }

170

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Issue Commands

The mongo shell provides a helper method for running commands called db.runCommand(). The following
operation in mongo runs the above command:
db.runCommand( { isMaster: 1 } )

Many drivers (page 95) provide an equivalent for the db.runCommand() method. Internally, running commands
with db.runCommand() is equivalent to a special query against the $cmd collection.
Many common commands have their own shell helpers or wrappers in the mongo shell and drivers, such as the
db.isMaster() method in the mongo JavaScript shell.
admin Database Commands

You must run some commands on the admin database. Normally, these operations resemble the followings:
use admin
db.runCommand( {buildInfo: 1} )

However, theres also a command helper that automatically runs the command in the context of the admin database:
db._adminCommand( {buildInfo: 1} )

Command Responses

All commands return, at minimum, a document with an ok field indicating whether the command has succeeded:
{ 'ok': 1 }

Failed commands return the ok field with a value of 0.

Manage mongod Processes
MongoDB runs as a standard program.
You can start MongoDB from a command line
by issuing the mongod command and specifying options.
For a list of options, see
http://docs.mongodb.org/manualreference/program/mongod. MongoDB can also run as a
Windows service. For details, see MongoDB as a Windows Service (page 15). To install MongoDB, see Install
MongoDB (page 3).
The following examples assume the directory containing the mongod process is in your system paths. The mongod
process is the primary database process that runs on an individual server. mongos provides a coherent MongoDB
interface equivalent to a mongod from the perspective of a client. The mongo binary provides the administrative
shell.
This document page discusses the mongod process; however, some portions of this document may be applicable to
mongos instances.
See also:
Run-time Database Configuration (page 146), http://docs.mongodb.org/manualreference/program/mongod,
http://docs.mongodb.org/manualreference/program/mongos,
and
http://docs.mongodb.org/manualreference/configuration-options.

4.2. Administration Tutorials

171

MongoDB Documentation, Release 2.4.9

Start mongod

By default, MongoDB stores data in the /data/db directory. On Windows, MongoDB stores data in C:\data\db.
On all platforms, MongoDB listens for connections from clients on port 27017.
To start MongoDB using all defaults, issue the following command at the system shell:
mongod

Specify a Data Directory If you want mongod to store data files at a path other than /data/db you can specify
a dbpath. The dbpath must exist before you start mongod. If it does not exist, create the directory and the
permissions so that mongod can read and write data to this path. For more information on permissions, see the
security operations documentation (page 238).
To specify a dbpath for mongod to use as a data directory, use the --dbpath option. The following invocation
will start a mongod instance and store data in the /srv/mongodb path
mongod --dbpath /srv/mongodb/

Specify a TCP Port Only a single process can listen for connections on a network interface at a time. If you run
multiple mongod processes on a single machine, or have other processes that must use this port, you must assign each
a different port to listen on for client connections.
To specify a port to mongod, use the --port option on the command line. The following command starts mongod
listening on port 12345:
mongod --port 12345

Use the default port number when possible, to avoid confusion.

Start mongod as a Daemon To run a mongod process as a daemon (i.e. fork), and write its output to a log file,
use the --fork and --logpath options. You must create the log directory; however, mongod will create the log
file if it does not exist.
The following command starts mongod as a daemon and records log output to /var/log/mongodb.log.
mongod --fork --logpath /var/log/mongodb.log

Additional Configuration Options For an overview of common configurations and common configuration deployments. configurations for common use cases, see Run-time Database Configuration (page 146).
Stop mongod

In a clean shutdown a mongod completes all pending operations, flushes all data to data files, and closes all data files.
Other shutdowns are unclean and can compromise the validity the data files.
To ensure a clean shutdown, always shutdown mongod instances using one of the following methods:
Use shutdownServer() Shut down the mongod from the mongo shell using the db.shutdownServer()
method as follows:
use admin
db.shutdownServer()

172

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Calling the same method from a control script accomplishes the same result.
For systems with auth enabled, users may only issue db.shutdownServer() when authenticated to the admin
database or via the localhost interface on systems without authentication enabled.
Use --shutdown From the Linux command line, shut down the mongod using the --shutdown option in the
following command:
mongod --shutdown

Use CTRL-C When running the mongod instance in interactive mode (i.e. without --fork), issue Control-C
to perform a clean shutdown.
Use kill From the Linux command line, shut down a specific mongod instance using the following command:
kill <mongod process ID>

Warning: Never use kill -9 (i.e. SIGKILL) to terminate a mongod instance.

Stop a Replica Set

Procedure If the mongod is the primary in a replica set, the shutdown process for these mongod instances has the
following steps:
1. Check how up-to-date the secondaries are.
2. If no secondary is within 10 seconds of the primary, mongod will return a message that it will not shut down.
You can pass the shutdown command a timeoutSecs argument to wait for a secondary to catch up.
3. If there is a secondary within 10 seconds of the primary, the primary will step down and wait for the secondary
to catch up.
4. After 60 seconds or once the secondary has caught up, the primary will shut down.
Force Replica Set Shutdown If there is no up-to-date secondary and you want the primary to shut down, issue the
shutdown command with the force argument, as in the following mongo shell operation:
db.adminCommand({shutdown : 1, force : true})

To keep checking the secondaries for a specified number of seconds if none are immediately up-to-date, issue
shutdown with the timeoutSecs argument. MongoDB will keep checking the secondaries for the specified
number of seconds if none are immediately up-to-date. If any of the secondaries catch up within the allotted time, the
primary will shut down. If no secondaries catch up, it will not shut down.
The following command issues shutdown with timeoutSecs set to 5:
db.adminCommand({shutdown : 1, timeoutSecs : 5})

Alternately you can use the timeoutSecs argument with the db.shutdownServer() method:
db.shutdownServer({timeoutSecs : 5})

4.2. Administration Tutorials

173

MongoDB Documentation, Release 2.4.9

Analyze Performance of Database Operations

The database profiler collects fine grained data about MongoDB write operations, cursors, database commands on
a running mongod instance. You can enable profiling on a per-database or per-instance basis. The profiling level
(page 174) is also configurable when enabling profiling.
The database profiler writes all the data it collects to the system.profile (page 227) collection, which is a capped
collection (page 161). See Database Profiler Output (page 230) for overview of the data in the system.profile
(page 227) documents created by the profiler.
This document outlines a number of key administration options for the database profiler. For additional related information, consider the following resources:
Database Profiler Output (page 230)
Profile Command
http://docs.mongodb.org/manualreference/method/db.currentOp
Profiling Levels

The following profiling levels are available:

0 - the profiler is off, does not collect any data. mongod always writes operations longer than the slowms
threshold to its log.
1 - collects profiling data for slow operations only. By default slow operations are those slower than 100
milliseconds.
You can modify the threshold for slow operations with the slowms runtime option or the setParameter
command. See the Specify the Threshold for Slow Operations (page 175) section for more information.
2 - collects profiling data for all database operations.
Enable Database Profiling and Set the Profiling Level

You can enable database profiling from the mongo shell or through a driver using the profile command. This
section will describe how to do so from the mongo shell. See your driver documentation (page 95) if you want to
control the profiler from within your application.
When you enable profiling, you also set the profiling level (page 174). The profiler records data in the
system.profile (page 227) collection. MongoDB creates the system.profile (page 227) collection in a
database after you enable profiling for that database.
To enable profiling and set the profiling level, use the db.setProfilingLevel() helper in the mongo shell,
passing the profiling level as a parameter. For example, to enable profiling for all database operations, consider the
following operation in the mongo shell:
db.setProfilingLevel(2)

The shell returns a document showing the previous level of profiling. The "ok" :
operation succeeded:

1 key-value pair indicates the

{ "was" : 0, "slowms" : 100, "ok" : 1 }

To verify the new setting, see the Check Profiling Level (page 175) section.

174

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Specify the Threshold for Slow Operations The threshold for slow operations applies to the entire mongod instance. When you change the threshold, you change it for all databases on the instance.
Important: Changing the slow operation threshold for the database profiler also affects the profiling subsystems
slow operation threshold for the entire mongod instance. Always set the threshold to the highest useful value.
By default the slow operation threshold is 100 milliseconds. Databases with a profiling level of 1 will log operations
slower than 100 milliseconds.
To change the threshold, pass two parameters to the db.setProfilingLevel() helper in the mongo shell. The
first parameter sets the profiling level for the current database, and the second sets the default slow operation threshold
for the entire mongod instance.
For example, the following command sets the profiling level for the current database to 0, which disables profiling,
and sets the slow-operation threshold for the mongod instance to 20 milliseconds. Any database on the instance with
a profiling level of 1 will use this threshold:
db.setProfilingLevel(0,20)

Check Profiling Level To view the profiling level (page 174), issue the following from the mongo shell:
db.getProfilingStatus()

The shell returns a document similar to the following:

{ "was" : 0, "slowms" : 100 }

The was field indicates the current level of profiling.

The slowms field indicates how long an operation must exist in milliseconds for an operation to pass the slow
threshold. MongoDB will log operations that take longer than the threshold if the profiling level is 1. This document
returns the profiling level in the was field. For an explanation of profiling levels, see Profiling Levels (page 174).
To return only the profiling level, use the db.getProfilingLevel() helper in the mongo as in the following:
db.getProfilingLevel()

Disable Profiling To disable profiling, use the following helper in the mongo shell:
db.setProfilingLevel(0)

Enable Profiling for an Entire mongod Instance For development purposes in testing environments, you can
enable database profiling for an entire mongod instance. The profiling level applies to all databases provided by the
mongod instance.
To enable profiling for a mongod instance, pass the following parameters to mongod at startup or within the
configuration file:
mongod --profile=1 --slowms=15

This sets the profiling level to 1, which collects profiling data for slow operations only, and defines slow operations as
those that last longer than 15 milliseconds.
See also:
profile and slowms.

4.2. Administration Tutorials

175

MongoDB Documentation, Release 2.4.9

Database Profiling and Sharding You cannot enable profiling on a mongos instance. To enable profiling in a
sharded cluster, you must enable profiling for each mongod instance in the cluster.
View Profiler Data

The database profiler logs information about database operations in the system.profile (page 227) collection.
To view profiling information, query the system.profile (page 227) collection. To view example queries, see
Example Profiler Data Queries (page 176)
For an explanation of the output data, see Database Profiler Output (page 230).
Example Profiler Data Queries This section displays example queries to the system.profile (page 227) collection. For an explanation of the query output, see Database Profiler Output (page 230).
To return the most recent 10 log entries in the system.profile (page 227) collection, run a query similar to the
following:
db.system.profile.find().limit(10).sort( { ts : -1 } ).pretty()

To return all operations except command operations ($cmd), run a query similar to the following:
db.system.profile.find( { op: { $ne : 'command' } } ).pretty()

To return operations for a particular collection, run a query similar to the following. This example returns operations
in the mydb databases test collection:
db.system.profile.find( { ns : 'mydb.test' } ).pretty()

To return operations slower than 5 milliseconds, run a query similar to the following:
db.system.profile.find( { millis : { $gt : 5 } } ).pretty()

To return information from a certain time range, run a query similar to the following:
db.system.profile.find(
{
ts : {
$gt : new ISODate("2012-12-09T03:00:00Z") ,
$lt : new ISODate("2012-12-09T03:40:00Z")
}
}
).pretty()

The following example looks at the time range, suppresses the user field from the output to make it easier to read,
and sorts the results by how long each operation took to run:
db.system.profile.find(
{
ts : {
$gt : new ISODate("2011-07-12T03:00:00Z") ,
$lt : new ISODate("2011-07-12T03:40:00Z")
}
},
{ user : 0 }
).sort( { millis : -1 } )

176

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Show the Five Most Recent Events On a database that has profiling enabled, the show profile helper in the
mongo shell displays the 5 most recent operations that took at least 1 millisecond to execute. Issue show profile
from the mongo shell, as follows:
show profile

Profiler Overhead

When enabled, profiling has a minor effect on performance. The system.profile (page 227) collection is a
capped collection with a default size of 1 megabyte. A collection of this size can typically store several thousand
profile documents, but some application may use more or less profiling data per operation.
To change the size of the system.profile (page 227) collection, you must:
1. Disable profiling.
2. Drop the system.profile (page 227) collection.
3. Create a new system.profile (page 227) collection.
4. Re-enable profiling.
For example, to create a new system.profile (page 227) collection thats 4000000 bytes, use the following
sequence of operations in the mongo shell:
db.setProfilingLevel(0)
db.system.profile.drop()
db.createCollection( "system.profile", { capped: true, size:4000000 } )
db.setProfilingLevel(1)

Change Size of system.profile Collection

To change the size of the system.profile (page 227) collection on a secondary, you
must stop the secondary, run it as a standalone, and then perform the steps above.
When
done, restart the standalone as a member of the replica set.
For more information, see
http://docs.mongodb.org/manualtutorial/perform-maintence-on-replica-set-members.
Monitor MongoDB with SNMP
New in version 2.2.
Enterprise Feature
This feature is only available in MongoDB Enterprise.
This document outlines the use and operation of MongoDBs SNMP extension, which is only available in MongoDB
Enterprise71 .
Prerequisites

Install MongoDB Enterprise MongoDB Enterprise

71 http://www.mongodb.com/products/mongodb-enterprise

4.2. Administration Tutorials

177

MongoDB Documentation, Release 2.4.9

Included Files The Enterprise packages contain the following files:

MONGO-MIB.txt:
The MIB file that describes the data (i.e. schema) for MongoDBs SNMP output
mongod.conf:
The SNMP configuration file for reading the SNMP output of MongoDB. The SNMP configures the community
names, permissions, access controls, etc.
Required Packages To use SNMP, you must install several prerequisites. The names of the packages vary by
distribution and are as follows:
Ubuntu 11.04 requires libssl0.9.8, snmp-mibs-downloader, snmp, and snmpd. Issue a command
such as the following to install these packages:
sudo apt-get install libssl0.9.8 snmp snmpd snmp-mibs-downloader

Red Hat Enterprise Linux 6.x series and Amazon Linux AMI require libssl, net-snmp,
net-snmp-libs, and net-snmp-utils. Issue a command such as the following to install these packages:
sudo yum install openssl net-snmp net-snmp-libs net-snmp-utils

SUSE Enterprise Linux requires libopenssl0_9_8, libsnmp15, slessp1-libsnmp15, and

snmp-mibs. Issue a command such as the following to install these packages:
sudo zypper install libopenssl0_9_8 libsnmp15 slessp1-libsnmp15 snmp-mibs

Configure SNMP

Install MIB Configuration Files

the following command:

Ensure that the MIB directory /usr/share/snmp/mibs exists. If not, issue

sudo mkdir -p /usr/share/snmp/mibs

Use the following command to create a symbolic link:

sudo ln -s <path>MONGO-MIB.txt /usr/share/snmp/mibs/

Replace [/path/to/mongodb/distribution/] with the path to your MONGO-MIB.txt configuration file.

Copy the mongod.conf file into the /etc/snmp directory with the following command:
cp mongod.conf /etc/snmp/mongod.conf

Start Up You can control MongoDB Enterprise using default or custom control scripts, just as with any other
mongod:
Use the following command to view all SNMP options available in your MongoDB:
mongod --help | grep snmp

The above command should return the following output:

Module snmp options:
--snmp-subagent
--snmp-master

178

run snmp subagent

run snmp as master

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Ensure that the following directories exist:

/data/db/ (This is the path where MongoDB stores the data files.)
/var/log/mongodb/ (This is the path where MongoDB writes the log output.)
If they do not, issue the following command:
mkdir -p /var/log/mongodb/ /data/db/

Start the mongod instance with the following command:

mongod --snmp-master --port 3001 --fork --dbpath /data/db/

--logpath /var/log/mongodb/1.log

Optionally, you can set these options in a configuration file.

To check if mongod is running with SNMP support, issue the following command:
ps -ef | grep 'mongod --snmp'

The command should return output that includes the following line. This indicates that the proper mongod instance is
running:
systemuser 31415 10260

0 Jul13 pts/16

00:00:00 mongod --snmp-master --port 3001 # [...]

Test SNMP Check for the snmp agent process listening on port 1161 with the following command:
sudo lsof -i :1161

which return the following output:

COMMAND PID
USER
mongod 9238 sysadmin

FD
10u

TYPE DEVICE SIZE/OFF NODE NAME

IPv4 96469
0t0 UDP localhost:health-polling

Similarly, this command:

netstat -anp | grep 1161

should return the following output:

udp

0 127.0.0.1:1161

0.0.0.0:*

9238/<path>/mongod

Run snmpwalk Locally snmpwalk provides tools for retrieving and parsing the SNMP data according to the
MIB. If you installed all of the required packages above, your system will have snmpwalk.
Issue the following command to collect data from mongod using SNMP:
snmpwalk -m MONGO-MIB -v 2c -c mongodb 127.0.0.1:1161 1.3.6.1.4.1.37601

You may also choose to specify the path to the MIB file:
snmpwalk -m /usr/share/snmp/mibs/MONGO-MIB -v 2c -c mongodb 127.0.0.1:1161 1.3.6.1.4.1.37601

Use this command only to ensure that you can retrieve and validate SNMP data from MongoDB.
Troubleshooting

Always check the logs for errors if something does not run as expected; see the log at /var/log/mongodb/1.log.
The presence of the following line indicates that the mongod cannot read the /etc/snmp/mongod.conf file:

4.2. Administration Tutorials

179

MongoDB Documentation, Release 2.4.9

[SNMPAgent] warning: error starting SNMPAgent as master err:1

Rotate Log Files

Overview

Log rotation using MongoDBs standard approach archives the current log file and starts a new one. To do this, the
mongod or mongos instance renames the current log file by appending a UTC (GMT) timestamp to the filename, in
ISODate format. It then opens a new log file, closes the old log file, and sends all new log entries to the new log file.
MongoDBs standard approach to log rotation only rotates logs in response to the logRotate command, or when
the mongod or mongos process receives a SIGUSR1 signal from the operating system.
Alternately, you may configure mongod to send log data to syslog. In this case, you can take advantage of alternate
logrotation tools.
See also:
For information on logging, see the Process Logging (page 142) section.
Log Rotation With MongoDB

The following steps create and rotate a log file:

1. Start a mongod with verbose logging, with appending enabled, and with the following log file:
mongod -v --logpath /var/log/mongodb/server1.log --logappend

2. In a separate terminal, list the matching files:

ls /var/log/mongodb/server1.log*

For results, you get:

server1.log

3. Rotate the log file using one of the following methods.

From the mongo shell, issue the logRotate command from the admin database:
use admin
db.runCommand( { logRotate : 1 } )

This is the only available method to rotate log files on Windows systems.
For Linux systems, rotate logs for a single process by issuing the following command:
kill -SIGUSR1 <mongod process id>

4. List the matching files again:

ls /var/log/mongodb/server1.log*

For results you get something similar to the following. The timestamps will be different.
server1.log

180

server1.log.2011-11-24T23-30-00

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

The example results indicate a log rotation performed at exactly 11:30 pm on November 24th, 2011
UTC, which is the local time offset by the local time zone. The original log file is the one with the timestamp.
The new log is server1.log file.
If you issue a second logRotate command an hour later, then an additional file would appear when listing
matching files, as in the following example:
server1.log

server1.log.2011-11-24T23-30-00

server1.log.2011-11-25T00-30-00

This operation does not modify the server1.log.2011-11-24T23-30-00 file created earlier, while
server1.log.2011-11-25T00-30-00 is the previous server1.log file, renamed. server1.log
is a new, empty file that receives all new log output.
Syslog Log Rotation

New in version 2.2.

To configure mongod to send log data to syslog rather than writing log data to a file, use the following procedure.
1. Start a mongod with the syslog option.
2. Store and rotate the log output using your systems default log rotation mechanism.
Important: You cannot use syslog with logpath.

Manage Journaling
MongoDB uses write ahead logging to an on-disk journal to guarantee write operation (page 42) durability and to
provide crash resiliency. Before applying a change to the data files, MongoDB writes the change operation to the
journal. If MongoDB should terminate or encounter an error before it can write the changes from the journal to the
data files, MongoDB can re-apply the write operation and maintain a consistent state.
Without a journal, if mongod exits unexpectedly, you must assume your data is in an inconsistent state, and you must
run either repair (page 203) or, preferably, resync (page 449) from a clean member of the replica set.
With journaling enabled, if mongod stops unexpectedly, the program can recover everything written to the journal,
and the data remains in a consistent state. By default, the greatest extent of lost writes, i.e., those not made to the
journal, are those made in the last 100 milliseconds. See journalCommitInterval for more information on the
default.
With journaling, if you want a data set to reside entirely in RAM, you need enough RAM to hold the data set plus
the write working set. The write working set is the amount of unique data you expect to see written between
re-mappings of the private view. For information on views, see Storage Views used in Journaling (page 234).
Important: Changed in version 2.0: For 64-bit builds of mongod, journaling is enabled by default. For other
platforms, see journal.

Procedures

Enable Journaling Changed in version 2.0: For 64-bit builds of mongod, journaling is enabled by default.
To enable journaling, start mongod with the --journal command line option.

4.2. Administration Tutorials

181

MongoDB Documentation, Release 2.4.9

If no journal files exist, when mongod starts, it must preallocate new journal files. During this operation, the mongod
is not listening for connections until preallocation completes: for some systems this may take a several minutes.
During this period your applications and the mongo shell are not available.

Disable Journaling

Warning: Do not disable journaling on production systems. If your mongod instance stops without shutti
down cleanly unexpectedly for any reason, (e.g. power failure) and you are not running with journaling, then y
must recover from an unaffected replica set member or backup, as described in repair (page 203).

To disable journaling, start mongod with the --nojournal command line option.
Get Commit Acknowledgment You can get commit acknowledgment with the getLastError command and the
j option. For details, see Write Concern Reference (page 83).
Avoid Preallocation Lag To avoid preallocation lag (page 234), you can preallocate files in the journal directory by
copying them from another instance of mongod.
Preallocated files do not contain data. It is safe to later remove them. But if you restart mongod with journaling,
mongod will create them again.
Example
The following sequence preallocates journal files for an instance of mongod running on port 27017 with a database
path of /data/db.
For demonstration purposes, the sequence starts by creating a set of journal files in the usual way.
1. Create a temporary directory into which to create a set of journal files:
mkdir ~/tmpDbpath

2. Create a set of journal files by staring a mongod instance that uses the temporary directory:
mongod --port 10000 --dbpath ~/tmpDbpath --journal

3. When you see the following log output, indicating mongod has the files, press CONTROL+C to stop the
mongod instance:
[initandlisten] waiting for connections on port 10000

4. Preallocate journal files for the new instance of mongod by moving the journal files from the data directory of
the existing instance to the data directory of the new instance:
mv ~/tmpDbpath/journal /data/db/

5. Start the new mongod instance:

mongod --port 27017 --dbpath /data/db --journal

Monitor Journal Status Use the following commands and methods to monitor journal status:
serverStatus
The serverStatus command returns database status information that is useful for assessing performance.

182

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

journalLatencyTest
Use journalLatencyTest to measure how long it takes on your volume to write to the disk in an appendonly fashion. You can run this command on an idle system to get a baseline sync time for journaling. You can
also run this command on a busy system to see the sync time on a busy system, which may be higher if the
journal directory is on the same volume as the data files.
The journalLatencyTest command also provides a way to check if your disk drive is buffering writes in
its local cache. If the number is very low (i.e., less than 2 milliseconds) and the drive is non-SSD, the drive
is probably buffering writes. In that case, enable cache write-through for the device in your operating system,
unless you have a disk controller card with battery backed RAM.
Change the Group Commit Interval Changed in version 2.0.
You can set the group commit interval using the --journalCommitInterval command line option. The allowed
range is 2 to 300 milliseconds.
Lower values increase the durability of the journal at the expense of disk performance.
Recover Data After Unexpected Shutdown On a restart after a crash, MongoDB replays all journal files in the
journal directory before the server becomes available. If MongoDB must replay journal files, mongod notes these
events in the log output.
There is no reason to run repairDatabase in these situations.
Store a JavaScript Function on the Server
Note: We do not recommend using server-side stored functions if possible.
There is a special system collection named system.js that can store JavaScript functions for reuse.
To store a function, you can use the db.collection.save(), as in the following example:
db.system.js.save(
{
_id : "myAddFunction" ,
value : function (x, y){ return x + y; }
}
);

The _id field holds the name of the function and is unique per database.
The value field holds the function definition
Once you save a function in the system.js collection, you can use the function from any JavaScript context (e.g.
eval command or the mongo shell method db.eval(), $where operator, mapReduce or mongo shell method
db.collection.mapReduce()).
Consider the following example from the mongo shell that first saves a function named echoFunction to the
system.js collection and calls the function using db.eval() method:
db.system.js.save(
{ _id: "echoFunction",
value : function(x) { return x; }
}
)
db.eval( "echoFunction( 'test' )" )

4.2. Administration Tutorials

183

MongoDB Documentation, Release 2.4.9

See http://github.com/mongodb/mongo/tree/master/jstests/storefunc.js for a full example.

New in version 2.1: In the mongo shell, you can use db.loadServerScripts() to load all the scripts saved in
the system.js collection for the current database. Once loaded, you can invoke the functions directly in the shell,
as in the following example:
db.loadServerScripts();
echoFunction(3);
myAddFunction(3, 5);

Upgrade to the Latest Revision of MongoDB

Revisions provide security patches, bug fixes, and new or changed features that do not contain any backward breaking
changes. Always upgrade to the latest revision in your release series. The third number in the MongoDB version
number (page 634) indicates the revision.
Before Upgrading

Ensure you have an up-to-date backup of your data set. See MongoDB Backup Methods (page 136).
Consult the following documents for any special considerations or compatibility issues specific to your MongoDB release:
The release notes, located at Release Notes (page 591).
The documentation for your driver. See MongoDB Drivers and Client Libraries (page 95).
If your installation includes replica sets, plan the upgrade during a predefined maintenance window.
Before you upgrade a production environment, use the procedures in this document to upgrade a staging environment that reproduces your production environment, to ensure that your production configuration is compatible
with all changes.
Upgrade Procedure

Important: Always backup all of your data before upgrading MongoDB.

Upgrade each mongod and mongos binary separately, using the procedure described here. When upgrading a binary,
use the procedure Upgrade a MongoDB Instance (page 185).
Follow this upgrade procedure:
1. For deployments that use authentication, first upgrade all of your MongoDB drivers (page 95). To upgrade, see
the documentation for your driver.
2. Upgrade sharded clusters, as described in Upgrade Sharded Clusters (page 185).
3. Upgrade any standalone instances. See Upgrade a MongoDB Instance (page 185).
4. Upgrade any replica sets that are not part of a sharded cluster, as described in Upgrade Replica Sets (page 186).

184

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Upgrade a MongoDB Instance

To upgrade a mongod or mongos instance, use one of the following approaches:

Upgrade the instance using the operating systems package management tool and the official MongoDB packages. This is the preferred approach. See Install MongoDB (page 3).
Upgrade the instance by replacing the existing binaries with new binaries. See Replace the Existing Binaries
(page 185).
Replace the Existing Binaries

Important: Always backup all of your data before upgrading MongoDB.

This section describes how to upgrade MongoDB by replacing the existing binaries. The preferred approach to an
upgrade is to use the operating systems package management tool and the official MongoDB packages, as described
in Install MongoDB (page 3).
To upgrade a mongod or mongos instance by replacing the existing binaries:
1. Download the binaries for the latest MongoDB revision from the MongoDB Download Page72 and store the
binaries in a temporary location. The binaries download as compressed files that uncompress to the directory
structure used by the MongoDB installation.
2. Shutdown the instance.
3. Replace the existing MongoDB binaries with the downloaded binaries.
4. Restart the instance.
Upgrade Sharded Clusters

To upgrade a sharded cluster:

1. Disable the clusters balancer, as described in Disable the Balancer (page 533).
2. Upgrade each mongos instance by following the instructions below in Upgrade a MongoDB Instance
(page 185). You can upgrade the mongos instances in any order.
3. Upgrade each mongod config server (page 488) individually starting with the last config server listed in your
mongos --configdb string and working backward. To keep the cluster online, make sure at least one config
server is always running. For each config server upgrade, follow the instructions below in Upgrade a MongoDB
Instance (page 185)
Example
Given the following config string:
mongos --configdb cfg0.example.net:27019,cfg1.example.net:27019,cfg2.example.net:27019

You would upgrade the config servers in the following order:

(a) cfg2.example.net
(b) cfg1.example.net
(c) cfg0.example.net
72 http://downloads.mongodb.org/

4.2. Administration Tutorials

185

MongoDB Documentation, Release 2.4.9

4. Upgrade each shard.

If a shard is a replica set, upgrade the shard using the procedure below titled Upgrade Replica Sets
(page 186).
If a shard is a standalone instance, upgrade the shard using the procedure below titled Upgrade a MongoDB
Instance (page 185).
5. Re-enable the balancer, as described in Enable the Balancer (page 533).
Upgrade Replica Sets

To upgrade a replica set, upgrade each member individually, starting with the secondaries and finishing with the
primary. Plan the upgrade during a predefined maintenance window.
Upgrade Secondaries Upgrade each secondary separately as follows:
1. Upgrade the secondarys mongod binary by following the instructions below in Upgrade a MongoDB Instance
(page 185).
2. After upgrading a secondary, wait for the secondary to recover to the SECONDARY state before upgrading the
next instance. To check the members state, issue rs.status() in the mongo shell.
The secondary may briefly go into STARTUP2 or RECOVERING. This is normal. Make sure to wait for the
secondary to fully recover to SECONDARY before you continue the upgrade.
Upgrade the Primary
1. Step down the primary to initiate the normal failover (page 397) procedure. Using one of the following:
The rs.stepDown() helper in the mongo shell.
The replSetStepDown database command.
During failover, the set cannot accept writes. Typically this takes 10-20 seconds. Plan the upgrade during a
predefined maintenance window.
Note: Stepping down the primary is preferable to directly shutting down the primary. Stepping down expedites
the failover procedure.
2. Once the primary has stepped down, call the rs.status() method from the mongo shell until you see that
another member has assumed the PRIMARY state.
3. Shut down the original primary and upgrade its instance by following the instructions below in Upgrade a
MongoDB Instance (page 185).
MongoDB Tutorials
This page lists the tutorials available as part of the MongoDB Manual. In addition to these documents, you can refer
to the introductory MongoDB Tutorial (page 19). If there is a process or pattern that you would like to see included
here, please open a Jira Case73 .
73 https://jira.mongodb.org/browse/DOCS

186

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Getting Started

Install MongoDB on Linux Systems (page 9)

Install MongoDB on Red Hat Enterprise, CentOS, Fedora, or Amazon Linux (page 4)
Install MongoDB on Debian (page 7)
Install MongoDB on Ubuntu (page 6)
Install MongoDB on OS X (page 11)
Install MongoDB on Windows (page 13)
Getting Started with MongoDB (page 19)
Generate Test Data (page 23)
Administration

Replica Sets
Deploy a Replica Set (page 420)
Convert a Standalone to a Replica Set (page 432)
Add Members to a Replica Set (page 433)
Remove Members from Replica Set (page 435)
Replace a Replica Set Member (page 437)
Adjust Priority for Replica Set Member (page 437)
Resync a Member of a Replica Set (page 449)
Deploy a Geographically Redundant Replica Set (page 425)
Change the Size of the Oplog (page 445)
Force a Member to Become Primary (page 447)
Change Hostnames in a Replica Set (page 457)
Add an Arbiter to Replica Set (page 431)
Convert a Secondary to an Arbiter (page 443)
Configure a Secondarys Sync Target (page 461)
Configure a Delayed Replica Set Member (page 441)
Configure a Hidden Replica Set Member (page 439)
Configure Non-Voting Replica Set Member (page 442)
Prevent Secondary from Becoming Primary (page 438)
Configure Replica Set Tag Sets (page 450)
Manage Chained Replication (page 456)
Reconfigure a Replica Set with Unavailable Members (page 454)
Recover Data after an Unexpected Shutdown (page 203)
Troubleshoot Replica Sets (page 461)

4.2. Administration Tutorials

187

MongoDB Documentation, Release 2.4.9

Sharding
Deploy a Sharded Cluster (page 507)
Convert a Replica Set to a Replicated Sharded Cluster (page 515)
Add Shards to a Cluster (page 514)
Remove Shards from an Existing Sharded Cluster (page 534)
Deploy Three Config Servers for Production Deployments (page 515)
Migrate Config Servers with the Same Hostname (page 524)
Migrate Config Servers with Different Hostnames (page 524)
Replace a Config Server (page 525)
Migrate a Sharded Cluster to Different Hardware (page 526)
Backup Cluster Metadata (page 529)
Backup a Small Sharded Cluster with mongodump (page 198)
Backup a Sharded Cluster with Filesystem Snapshots (page 199)
Backup a Sharded Cluster with Database Dumps (page 200)
Restore a Single Shard (page 202)
Restore a Sharded Cluster (page 202)
Schedule Backup Window for Sharded Clusters (page 202)
Manage Shard Tags (page 541)
Basic Operations
Use Database Commands (page 170)
Recover Data after an Unexpected Shutdown (page 203)
Expire Data from Collections by Setting TTL (page 163)
Analyze Performance of Database Operations (page 174)
Rotate Log Files (page 180)
Build Old Style Indexes (page 346)
Manage mongod Processes (page 171)
Back Up and Restore with MongoDB Tools (page 195)
Backup and Restore with Filesystem Snapshots (page 190)
Security
Configure Linux iptables Firewall for MongoDB (page 245)
Configure Windows netsh Firewall for MongoDB (page 249)
Enable Authentication (page 257)
Create a User Administrator (page 258)
Add a User to a Database (page 259)
Generate a Key File (page 261)
188

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Deploy MongoDB with Kerberos Authentication (page 261)

Create a Vulnerability Report (page 265)
Development Patterns

Perform Two Phase Commits (page 69)

Isolate Sequence of Operations (page 76)
Create an Auto-Incrementing Sequence Field (page 78)
Enforce Unique Keys for Sharded Collections (page 542)
Aggregation Examples (page 292)
Model Data to Support Keyword Search (page 120)
Limit Number of Elements in an Array after an Update (page 81)
Perform Incremental Map-Reduce (page 302)
Troubleshoot the Map Function (page 304)
Troubleshoot the Reduce Function (page 305)
Store a JavaScript Function on the Server (page 183)
Text Search Patterns

Enable Text Search (page 359)

Create a text Index (page 359)
Search String Content for Text (page 360)
Specify a Language for Text Index (page 363)
Create text Index with Long Name (page 364)
Control Search Results with Weights (page 365)
Create text Index to Satisfy the filter Component of Text Search (page 367)
Limit the Number of Entries Scanned (page 366)
Data Modeling Patterns

Model One-to-One Relationships with Embedded Documents (page 106)

Model One-to-Many Relationships with Embedded Documents (page 107)
Model One-to-Many Relationships with Document References (page 108)
Model Data for Atomic Operations (page 119)
Model Tree Structures with Parent References (page 112)
Model Tree Structures with Child References (page 113)
Model Tree Structures with Materialized Paths (page 116)
Model Tree Structures with Nested Sets (page 118)

4.2. Administration Tutorials

189

MongoDB Documentation, Release 2.4.9

4.2.2 Backup and Recovery

The following tutorials describe backup and restoration for a mongod instance:
Backup and Restore with Filesystem Snapshots (page 190) An outline of procedures for creating MongoDB data set
backups using system-level file snapshot tool, such as LVM or native storage appliance tools.
Restore a Replica Set from MongoDB Backups (page 194) Describes procedure for restoring a replica set from an
archived backup such as a mongodump or MMS Backup74 file.
Back Up and Restore with MongoDB Tools (page 195) The procedure for writing the contents of a database to a
BSON (i.e. binary) dump file for backing up MongoDB databases.
Backup and Restore Sharded Clusters (page 198) Detailed procedures and considerations for backing up sharded
clusters and single shards.
Recover Data after an Unexpected Shutdown (page 203) Recover data from MongoDB data files that were not properly closed or have an invalid state.
Backup and Restore with Filesystem Snapshots
This document describes a procedure for creating backups of MongoDB systems using system-level tools, such as
LVM or storage appliance, as well as the corresponding restoration strategies.
These filesystem snapshots, or block-level backup methods use system level tools to create copies of the device that
holds MongoDBs data files. These methods complete quickly and work reliably, but require more system configuration outside of MongoDB.
See also:
MongoDB Backup Methods (page 136) and Back Up and Restore with MongoDB Tools (page 195).
Snapshots Overview

Snapshots work by creating pointers between the live data and a special snapshot volume. These pointers are theoretically equivalent to hard links. As the working data diverges from the snapshot, the snapshot process uses a
copy-on-write strategy. As a result the snapshot only stores modified data.
After making the snapshot, you mount the snapshot image on your file system and copy data from the snapshot. The
resulting backup contains a full copy of all data.
Snapshots have the following limitations:
The database must be valid when the snapshot takes place. This means that all writes accepted by the database
need to be fully written to disk: either to the journal or to data files.
If all writes are not on disk when the backup occurs, the backup will not reflect these changes. If writes are in
progress when the backup occurs, the data files will reflect an inconsistent state. With journaling all data-file
states resulting from in-progress writes are recoverable; without journaling you must flush all pending writes
to disk before running the backup operation and must ensure that no writes occur during the entire backup
procedure.
If you do use journaling, the journal must reside on the same volume as the data.
Snapshots create an image of an entire disk image. Unless you need to back up your entire system, consider
isolating your MongoDB data files, journal (if applicable), and configuration on one logical disk that doesnt
contain any other data.
74 http://mms.mongodb.com

190

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Alternately, store all MongoDB data files on a dedicated device so that you can make backups without duplicating extraneous data.
Ensure that you copy data from snapshots and onto other systems to ensure that data is safe from site failures.
Although different snapshots methods provide different capability, the LVM method outlined below does not
provide any capacity for capturing incremental backups.
Snapshots With Journaling If your mongod instance has journaling enabled, then you can use any kind of file
system or volume/block level snapshot tool to create backups.
If you manage your own infrastructure on a Linux-based system, configure your system with LVM to provide your disk
packages and provide snapshot capability. You can also use LVM-based setups within a cloud/virtualized environment.
Note: Running LVM provides additional flexibility and enables the possibility of using snapshots to back up MongoDB.

Snapshots with Amazon EBS in a RAID 10 Configuration If your deployment depends on Amazons Elastic
Block Storage (EBS) with RAID configured within your instance, it is impossible to get a consistent state across all
disks using the platforms snapshot tool. As an alternative, you can do one of the following:
Flush all writes to disk and create a write lock to ensure consistent state during the backup process.
If you choose this option see Create Backups on Instances that do not have Journaling Enabled (page 193).
Configure LVM to run and hold your MongoDB data files on top of the RAID within your system.
If you choose this option, perform the LVM backup operation described in Create a Snapshot (page 191).
Backup and Restore Using LVM on a Linux System

This section provides an overview of a simple backup process using LVM on a Linux system. While the tools, commands, and paths may be (slightly) different on your system the following steps provide a high level overview of the
backup operation.
Note: Only use the following procedure as a guideline for a backup system and infrastructure. Production backup
systems must consider a number of application specific requirements and factors unique to specific environments.

Create a Snapshot To create a snapshot with LVM, issue a command as root in the following format:
lvcreate --size 100M --snapshot --name mdb-snap01 /dev/vg0/mongodb

This command creates an LVM snapshot (with the --snapshot option) named mdb-snap01 of the mongodb
volume in the vg0 volume group.

This example creates a snapshot named mdb-snap01 located at http://docs.mongodb.org/manualdev/vg0/mdb-snap01

The location and paths to your systems volume groups and devices may vary slightly depending on your operating
systems LVM configuration.
The snapshot has a cap of at 100 megabytes, because of the parameter --size 100M. This size does not
reflect the total amount of the data on the disk, but rather the quantity of differences between the current
state of http://docs.mongodb.org/manualdev/vg0/mongodb and the creation of the snapshot (i.e.
http://docs.mongodb.org/manualdev/vg0/mdb-snap01.)

4.2. Administration Tutorials

191

MongoDB Documentation, Release 2.4.9

Warning: Ensure that you create snapshots with enough space to account for data growth, particularly for the
period of time that it takes to copy data out of the system or to a temporary image.
If your snapshot runs out of space, the snapshot image becomes unusable. Discard this logical volume and create
another.
The snapshot will exist when the command returns. You can restore directly from the snapshot at any time or by
creating a new logical volume and restoring from this snapshot to the alternate image.
While snapshots are great for creating high quality backups very quickly, they are not ideal as a format for storing
backup data. Snapshots typically depend and reside on the same storage infrastructure as the original disk images.
Therefore, its crucial that you archive these snapshots and store them elsewhere.
Archive a Snapshot After creating a snapshot, mount the snapshot and move the data to separate storage. Your
system might try to compress the backup images as you move the offline. The following procedure fully archives the
data from the snapshot:
umount /dev/vg0/mdb-snap01
dd if=/dev/vg0/mdb-snap01 | gzip > mdb-snap01.gz

The above command sequence does the following:

Ensures that the http://docs.mongodb.org/manualdev/vg0/mdb-snap01 device is not mounted.
Performs a block level copy of the entire snapshot image using the dd command and compresses the result in a
gzipped file in the current working directory.
Warning: This command will create a large gz file in your current working directory. Make sure that you
run this command in a file system that has enough free space.

Restore a Snapshot
mands:

To restore a snapshot created with the above method, issue the following sequence of com-

lvcreate --size 1G --name mdb-new vg0

gzip -d -c mdb-snap01.gz | dd of=/dev/vg0/mdb-new
mount /dev/vg0/mdb-new /srv/mongodb

The above sequence does the following:

Creates a new logical volume named mdb-new, in the http://docs.mongodb.org/manualdev/vg0
volume group. The path to the new device will be http://docs.mongodb.org/manualdev/vg0/mdb-new.
Warning: This volume will have a maximum size of 1 gigabyte. The original file system must have had a
total size of 1 gigabyte or smaller, or else the restoration will fail.
Change 1G to your desired volume size.
Uncompresses and unarchives the mdb-snap01.gz into the mdb-new disk image.
Mounts the mdb-new disk image to the /srv/mongodb directory. Modify the mount point to correspond to
your MongoDB data file location, or other location as needed.
Note: The restored snapshot will have a stale mongod.lock file. If you do not remove this file from the snapshot,
and MongoDB may assume that the stale lock file indicates an unclean shutdown. If youre running with journal
enabled, and you do not use db.fsyncLock(), you do not need to remove the mongod.lock file. If you use
db.fsyncLock() you will need to remove the lock.

192

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Restore Directly from a Snapshot

sequence of commands:

To restore a backup without writing to a compressed gz file, use the following

umount /dev/vg0/mdb-snap01
lvcreate --size 1G --name mdb-new vg0
dd if=/dev/vg0/mdb-snap01 of=/dev/vg0/mdb-new
mount /dev/vg0/mdb-new /srv/mongodb

Remote Backup Storage

You can implement off-system backups using the combined process (page 193) and SSH.

This sequence is identical to procedures explained above, except that it archives and compresses the backup on a
remote system using SSH.
Consider the following procedure:
umount /dev/vg0/mdb-snap01
dd if=/dev/vg0/mdb-snap01 | ssh [email protected] gzip > /opt/backup/mdb-snap01.gz
lvcreate --size 1G --name mdb-new vg0
ssh [email protected] gzip -d -c /opt/backup/mdb-snap01.gz | dd of=/dev/vg0/mdb-new
mount /dev/vg0/mdb-new /srv/mongodb

Create Backups on Instances that do not have Journaling Enabled

If your mongod instance does not run with journaling enabled, or if your journal is on a separate volume, obtaining a
functional backup of a consistent state is more complicated. As described in this section, you must flush all writes to
disk and lock the database to prevent writes during the backup process. If you have a replica set configuration, then
for your backup use a secondary which is not receiving reads (i.e. hidden member).
1. To flush writes to disk and to lock the database (to prevent further writes), issue the db.fsyncLock()
method in the mongo shell:
db.fsyncLock();

2. Perform the backup operation described in Create a Snapshot (page 191).

3. To unlock the database after the snapshot has completed, use the following command in the mongo shell:
db.fsyncUnlock();

Note: Changed in version 2.0: MongoDB 2.0 added db.fsyncLock() and db.fsyncUnlock() helpers
to the mongo shell. Prior to this version, use the fsync command with the lock option, as follows:
db.runCommand( { fsync: 1, lock: true } );
db.runCommand( { fsync: 1, lock: false } );

The database cannot be locked with db.fsyncLock() while profiling is enabled. You must disable profiling
before locking the database with db.fsyncLock(). Disable profiling using db.setProfilingLevel()
as follows in the mongo shell:
db.setProfilingLevel(0)

Warning: Changed in version 2.2: When used in combination with fsync or db.fsyncLock(),
mongod may block some reads, including those from mongodump, when queued write operation waits
behind the fsync lock.

4.2. Administration Tutorials

193

MongoDB Documentation, Release 2.4.9

Restore a Replica Set from MongoDB Backups

This procedure outlines the process for taking MongoDB data and restoring that data into a new replica set. Use this
approach for seeding test deployments from production backups as well as part of disaster recovery.
You cannot restore a single data set to three new mongod instances and then create a replica set. In this situation
MongoDB will force the secondaries to perform an initial sync. The procedures in this document describe the correct
and efficient ways to deploy a replica set.
Restore Database into a Single Node Replica Set

1. Obtain backup MongoDB Database files. These files may come from a file system snapshot. The
MongoDB Management Service (MMS)75 produces MongoDB database files for stored snapshots76 and point
and time snapshots77 . You can also use mongorestore to restore database files using data created with
mongodump. See Back Up and Restore with MongoDB Tools (page 195) for more information.
2. Start a mongod using data files from the backup as the dbpath. In the following example, /data/db is the
dbpath to the data files:
mongod --dbpath /data/db

3. Convert your standalone mongod process to a single node replica set by shutting down the mongod instance,
and restarting it with the --replSet option, as in the following example:
mongod --dbpath /data/db --replSet <replName>

Optional
Consider explicitly setting a oplogSize to control the size of the oplog created for this replica set member.
4. Connect to the mongod instance.
5. Use rs.initiate() to initiate the new replica set.
Add Members to the Replica Set

MongoDB provides two options for restoring secondary members of a replica set:
1. Manually copy the database files to each data directory.
2. Allow initial sync (page 412) to distribute data automatically.
The following sections outlines both approaches.
Note: If your database is large, initial sync can take a long time to complete. For large databases, it might be
preferable to copy the database files onto each host.

Copy Database Files and Restart mongod Instance Use the following sequence of operations to seed additional
members of the replica set with the restored data by copying MongoDB data files directly.
1. Shut down the mongod instance that you restored. Using --shutdown or db.shutdownServer() to
ensure a clean shut down.
75 https://mms.mongodb.com/?pk_campaign=mongodb-docs-restore-rs-tutorial
76 https://mms.mongodb.com/help/backup/tutorial/restore-from-snapshot/
77 https://mms.mongodb.com/help/backup/tutorial/restore-from-point-in-time-snapshot/

194

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

2. Copy the primarys data directory into the dbpath of the other members of the replica set. The dbpath is
/data/db by default.
3. Start the mongod instance that you restored.
4. In a mongo shell connected to the primary, add the secondaries to the replica set using rs.add(). See Deploy
a Replica Set (page 420) for more information about deploying a replica set.
Update Secondaries using Initial Sync Use the following sequence of operations to seed additional members of
the replica set with the restored data using the default initial sync operation.
1. Ensure that the data directories on the prospective replica set members are empty.
2. Add each prospective member to the replica set. Initial Sync (page 412) will copy the data from the primary to
the other members of the replica set.
Back Up and Restore with MongoDB Tools
This document describes the process for writing and restoring backups to files in binary format with the mongodump
and mongorestore tools.
Use these tools for backups if other backup methods, such as the MMS Backup Service78 or file system snapshots
(page 190) are unavailable.
See also:
MongoDB Backup Methods (page 136), http://docs.mongodb.org/manualreference/program/mongodump,
and http://docs.mongodb.org/manualreference/program/mongorestore.
Backup a Database with mongodump

Important: mongodump does not dump the content of the local database.

Basic mongodump Operations The mongodump utility can back up data by either:
connecting to a running mongod or mongos instance, or
accessing data files without an active instance.
The utility can create a backup for an entire server, database or collection, or can use a query to backup just part of a
collection.
When you run mongodump without any arguments, the command connects to the MongoDB instance on the local
system (e.g. 127.0.0.1 or localhost) on port 27017 and creates a database backup named dump/ in the
current directory.
To backup data from a mongod or mongos instance running on the same machine and on the default port of 27017
use the following command:
mongodump

Warning: The data format used by mongodump from version 2.2 or later is incompatible with earlier versions
of mongod. Do not use recent versions of mongodump to back up older data stores.
78 https://mms.mongodb.com/?pk_campaign=mongodb-docs-tools

4.2. Administration Tutorials

195

MongoDB Documentation, Release 2.4.9

To limit the amount of data included in the database dump, you can specify --db and --collection as options to
the mongodump command. For example:
mongodump --dbpath /data/db/ --out /data/backup/
mongodump --host mongodb.example.net --port 27017

mongodump will write BSON files that hold a copy of data accessible via the mongod listening on port 27017 of
the mongodb.example.net host.
mongodump --collection collection --db test

This command creates a dump of the collection named collection from the database test in a dump/ subdirectory of the current working directory.
Point in Time Operation Using Oplogs Use the --oplog option with mongodump to collect the oplog entries to
build a point-in-time snapshot of a database within a replica set. With --oplog, mongodump copies all the data from
the source database as well as all of the oplog entries from the beginning of the backup procedure to until the backup
procedure completes. This backup procedure, in conjunction with mongorestore --oplogReplay, allows you
to restore a backup that reflects the specific moment in time that corresponds to when mongodump completed creating
the dump file.
Create Backups Without a Running mongod Instance If your MongoDB instance is not running, you can use
the --dbpath option to specify the location to your MongoDB instances database files. mongodump reads from
the data files directly with this operation. This locks the data directory to prevent conflicting writes. The mongod
process must not be running or attached to these data files when you run mongodump in this configuration. Consider
the following example:
Example
Backup a MongoDB Instance Without a Running mongod
Given a MongoDB instance that contains the customers, products, and suppliers databases, the following mongodump operation backs up the databases using the --dbpath option, which specifies the location of the
database files on the host:
mongodump --dbpath /data -o dataout

The --out option allows you to specify the directory where mongodump will save the backup. mongodump creates
a separate backup directory for each of the backed up databases: dataout/customers, dataout/products,
and dataout/suppliers.

Create Backups from Non-Local mongod Instances The --host and --port options for mongodump allow
you to connect to and backup from a remote host. Consider the following example:

mongodump --host mongodb1.example.net --port 3017 --username user --password pass --out /opt/backup/m

On any mongodump command you may, as above, specify username and password credentials to specify database
authentication.
Restore a Database with mongorestore

The mongorestore utility restores a binary backup created by mongodump. By default, mongorestore looks
for a database backup in the dump/ directory.
196

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

The mongorestore utility can restore data either by:

connecting to a running mongod or mongos directly, or
writing to a set of MongoDB data files without use of a running mongod.
mongorestore can restore either an entire database backup or a subset of the backup.
To use mongorestore to connect to an active mongod or mongos, use a command with the following prototype
form:
mongorestore --port <port number> <path to the backup>

To use mongorestore to write to data files without using a running mongod, use a command with the following
prototype form:
mongorestore --dbpath <database path> <path to the backup>

Consider the following example:

mongorestore dump-2012-10-25/

Here, mongorestore imports the database backup in the dump-2012-10-25 directory to the mongod instance
running on the localhost interface.
Restore Point in Time Oplog Backup If you created your database dump using the --oplog option to ensure a
point-in-time snapshot, call mongorestore with the --oplogReplay option, as in the following example:
mongorestore --oplogReplay

You may also consider using the mongorestore --objcheck option to check the integrity of objects while
inserting them into the database, or you may consider the mongorestore --drop option to drop each collection
from the database before restoring from backups.
Restore a Subset of data from a Binary Database Dump mongorestore also includes the ability to a filter to
all input before inserting it into the new database. Consider the following example:
mongorestore --filter '{"field": 1}'

Here, mongorestore only adds documents to the database from the dump located in the dump/ folder if the
documents have a field name field that holds a value of 1. Enclose the filter in single quotes (e.g. ) to prevent the
filter from interacting with your shell environment.
Restore Without a Running mongod mongorestore can write data to MongoDB data files without needing to
connect to a mongod directly.
Example
Restore a Database Without a Running mongod
Given a set of backed up databases in the /data/backup/ directory:
/data/backup/customers,
/data/backup/products, and
/data/backup/suppliers
The following mongorestore command restores the products database. The command uses the --dbpath
option to specify the path to the MongoDB data files:

4.2. Administration Tutorials

197

MongoDB Documentation, Release 2.4.9

mongorestore --dbpath /data/db --journal /data/backup/products

The mongorestore imports the database backup in the /data/backup/products directory to the mongod
instance that runs on the localhost interface. The mongorestore operation imports the backup even if the mongod
is not running.
The --journal option ensures that mongorestore records all operation in the durability journal. The journal
prevents data file corruption if anything (e.g. power failure, disk failure, etc.) interrupts the restore operation.
See also:
http://docs.mongodb.org/manualreference/program/mongodump
http://docs.mongodb.org/manualreference/program/mongorestore.

and

Restore Backups to Non-Local mongod Instances By default, mongorestore connects to a MongoDB instance
running on the localhost interface (e.g. 127.0.0.1) and on the default port (27017). If you want to restore to a
different host or port, use the --host and --port options.
Consider the following example:

mongorestore --host mongodb1.example.net --port 3017 --username user --password pass /opt/backup/mong

As above, you may specify username and password connections if your mongod requires authentication.
Backup and Restore Sharded Clusters
The following tutorials describe backup and restoration for sharded clusters:
Backup a Small Sharded Cluster with mongodump (page 198) If your sharded cluster holds a small data set, you
can use mongodump to capture the entire backup in a reasonable amount of time.
Backup a Sharded Cluster with Filesystem Snapshots (page 199) Use file system snapshots back up each component in the sharded cluster individually. The procedure involves stopping the cluster balancer. If your system
configuration allows file system backups, this might be more efficient than using MongoDB tools.
Backup a Sharded Cluster with Database Dumps (page 200) Create backups using mongodump to back up each
component in the cluster individually.
Schedule Backup Window for Sharded Clusters (page 202) Limit the operation of the cluster balancer to provide a
window for regular backup operations.
Restore a Single Shard (page 202) An outline of the procedure and consideration for restoring a single shard from a
backup.
Restore a Sharded Cluster (page 202) An outline of the procedure and consideration for restoring an entire sharded
cluster from backup.
Backup a Small Sharded Cluster with mongodump

Overview If your sharded cluster holds a small data set, you can connect to a mongos using mongodump. You can
create backups of your MongoDB cluster, if your backup infrastructure can capture the entire backup in a reasonable
amount of time and if you have a storage system that can hold the complete MongoDB data set.
See MongoDB Backup Methods (page 136) and Backup and Restore Sharded Clusters (page 198) for a complete
information on backups in MongoDB and backups of sharded clusters in particular.
Important: By default mongodump issue its queries to the non-primary nodes.
198

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Considerations If you use mongodump without specifying a database or collection, mongodump will capture
collection data and the cluster meta-data from the config servers (page 488).
You cannot use the --oplog option for mongodump when capturing data from mongos. As a result, if you need
to capture a backup that reflects a single moment in time, you must stop all writes to the cluster for the duration of the
backup operation.
Procedure
Capture Data You can perform a backup of a sharded cluster by connecting mongodump to a mongos. Use the
following operation at your systems prompt:
mongodump --host mongos3.example.net --port 27017

mongodump will write BSON files that hold a copy of data stored in the sharded cluster accessible via the mongos
listening on port 27017 of the mongos3.example.net host.
Restore Data Backups created with mongodump do not reflect the chunks or the distribution of data in the sharded
collection or collections. Like all mongodump output, these backups contain separate directories for each database
and BSON files for each collection in that database.
You can restore mongodump output to any MongoDB instance, including a standalone, a replica set, or a new sharded
cluster. When restoring data to sharded cluster, you must deploy and configure sharding before restoring data from
the backup. See Deploy a Sharded Cluster (page 507) for more information.
Backup a Sharded Cluster with Filesystem Snapshots

Overview This document describes a procedure for taking a backup of all components of a sharded cluster. This procedure uses file system snapshots to capture a copy of the mongod instance. An alternate procedure uses mongodump
to create binary database dumps when file-system snapshots are not available. See Backup a Sharded Cluster with
Database Dumps (page 200) for the alternate procedure.
See MongoDB Backup Methods (page 136) and Backup and Restore Sharded Clusters (page 198) for a complete
information on backups in MongoDB and backups of sharded clusters in particular.
Important: To capture a point-in-time backup from a sharded cluster you must stop all writes to the cluster. On a
running production system, you can only capture an approximation of point-in-time snapshot.

Procedure In this procedure, you will stop the cluster balancer and take a backup up of the config database, and
then take backups of each shard in the cluster using a file-system snapshot tool. If you need an exact moment-in-time
snapshot of the system, you will need to stop all application writes before taking the filesystem snapshots; otherwise
the snapshot will only approximate a moment in time.
For approximate point-in-time snapshots, you can improve the quality of the backup while minimizing impact on the
cluster by taking the backup from a secondary member of the replica set that provides each shard.
1. Disable the balancer process that equalizes the distribution of data among the shards. To disable the balancer,
use the sh.stopBalancer() method in the mongo shell. For example:
use config
sh.setBalancerState(false)

4.2. Administration Tutorials

199

MongoDB Documentation, Release 2.4.9

For more information, see the Disable the Balancer (page 533) procedure.
Warning: It is essential that you stop the balancer before creating backups. If the balancer remains active,
your resulting backups could have duplicate data or miss some data, as chunks may migrate while recording
backups.
2. Lock one secondary member of each replica set in each shard so that your backups reflect the state of your
database at the nearest possible approximation of a single moment in time. Lock these mongod instances in as
short of an interval as possible.
To lock a secondary, connect through the mongo shell to the secondary members mongod instance and issue
the db.fsyncLock() method.
3. Back up one of the config servers (page 488). Backing up a config server backs up the sharded clusters metadata.
You need back up only one config server, as they all hold the same data
Do one of the following to back up one of the config servers:
Create a file-system snapshot of the config server. Use the procedure in Backup and Restore with Filesystem
Snapshots (page 190).
Important:
This is only avalible if the config server has journal is enabled.
db.fsyncLock() on config databases.

Never use

Use mongodump to backup the config server. Issue mongodump against one of the config mongod
instances or via the mongos.
If you are running MongoDB 2.4 or later with the --configsvr option, then include the --oplog
option when running mongodump to ensure that the dump includes a partial oplog containing operations
from the duration of the mongodump operation. For example:
mongodump --oplog --db config

4. Back up the replica set members of the shards that you locked. You may back up the shards in parallel. For each
shard, create a snapshot. Use the procedure in Backup and Restore with Filesystem Snapshots (page 190).
5. Unlock all locked replica set members of each shard using the db.fsyncUnlock() method in the mongo
shell.
6. Re-enable the balancer with the sh.setBalancerState() method.
Use the following command sequence when connected to the mongos with the mongo shell:
use config
sh.setBalancerState(true)

Backup a Sharded Cluster with Database Dumps

Overview This document describes a procedure for taking a backup of all components of a sharded cluster. This
procedure uses mongodump to create dumps of the mongod instance. An alternate procedure uses file system snapshots to capture the backup data, and may be more efficient in some situations if your system configuration allows file
system backups. See Backup a Sharded Cluster with Filesystem Snapshots (page 199).
See MongoDB Backup Methods (page 136) and Backup and Restore Sharded Clusters (page 198) for a complete
information on backups in MongoDB and backups of sharded clusters in particular.
Important: To capture a point-in-time backup from a sharded cluster you must stop all writes to the cluster. On a
running production system, you can only capture an approximation of point-in-time snapshot.

200

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Procedure In this procedure, you will stop the cluster balancer and take a backup up of the config database, and then
take backups of each shard in the cluster using mongodump to capture the backup data. If you need an exact momentin-time snapshot of the system, you will need to stop all application writes before taking the filesystem snapshots;
otherwise the snapshot will only approximate a moment of time.
For approximate point-in-time snapshots, you can improve the quality of the backup while minimizing impact on the
cluster by taking the backup from a secondary member of the replica set that provides each shard.
1. Disable the balancer process that equalizes the distribution of data among the shards. To disable the balancer,
use the sh.stopBalancer() method in the mongo shell. For example:
use config
sh.setBalancerState(false)

For more information, see the Disable the Balancer (page 533) procedure.
Warning: It is essential that you stop the balancer before creating backups. If the balancer remains active,
your resulting backups could have duplicate data or miss some data, as chunks migrate while recording
backups.
2. Lock one member of each replica set in each shard so that your backups reflect the state of your database at
the nearest possible approximation of a single moment in time. Lock these mongod instances in as short of an
interval as possible.
To lock or freeze a sharded cluster, you shut down one member of each replica set. Ensure that the oplog has
sufficient capacity to allow these secondaries to catch up to the state of the primaries after finishing the backup
procedure. See Oplog Size (page 411) for more information.
3. Use mongodump to backup one of the config servers (page 488). This backs up the clusters metadata. You
only need to back up one config server, as they all hold the same data.
Use the mongodump tool to capture the content of the config mongod instances.
Your config servers must run MongoDB 2.4 or later with the --configsvr option and the mongodump
option must include the --oplog to capture a consistent copy of the config database:
mongodump --oplog --db config

4. Back up the replica set members of the shards that shut down using mongodump and specifying the --dbpath
option. You may back up the shards in parallel. Consider the following invocation:
mongodump --journal --dbpath /data/db/ --out /data/backup/

You must run this command on the system where the mongod ran. This operation will use journaling and create
a dump of the entire mongod instance with data files stored in /data/db/. mongodump will write the output
of this dump to the /data/backup/ directory.
5. Restart all stopped replica set members of each shard as normal and allow them to catch up with the state of the
primary.
6. Re-enable the balancer with the sh.setBalancerState() method.
Use the following command sequence when connected to the mongos with the mongo shell:
use config
sh.setBalancerState(true)

4.2. Administration Tutorials

201

MongoDB Documentation, Release 2.4.9

Schedule Backup Window for Sharded Clusters

Overview In a sharded cluster, the balancer process is responsible for distributing sharded data around the cluster,
so that each shard has roughly the same amount of data.
However, when creating backups from a sharded cluster it is important that you disable the balancer while taking
backups to ensure that no chunk migrations affect the content of the backup captured by the backup procedure. Using
the procedure outlined in the section Disable the Balancer (page 533) you can manually stop the balancer process
temporarily. As an alternative you can use this procedure to define a balancing window so that the balancer is always
disabled during your automated backup operation.
Procedure If you have an automated backup schedule, you can disable all balancing operations for a period of time.
For instance, consider the following command:
use config
db.settings.update( { _id : "balancer" }, { $set : { activeWindow : { start : "6:00", stop : "23:00"

This operation configures the balancer to run between 6:00am and 11:00pm, server time. Schedule your backup
operation to run and complete outside of this time. Ensure that the backup can complete outside the window when
the balancer is running and that the balancer can effectively balance the collection among the shards in the window
allotted to each.
Restore a Single Shard

Overview Restoring a single shard from backup with other unaffected shards requires a number of special considerations and practices. This document outlines the additional tasks you must perform when restoring a single shard.
Consider the following resources on backups in general as well as backup and restoration of sharded clusters specifically:
Backup and Restore Sharded Clusters (page 198)
Restore a Sharded Cluster (page 202)
MongoDB Backup Methods (page 136)
Procedure Always restore sharded clusters as a whole. When you restore a single shard, keep in mind that the
balancer process might have moved chunks to or from this shard since the last backup. If thats the case, you must
manually move those chunks, as described in this procedure.
1. Restore the shard as you would any other mongod instance. See MongoDB Backup Methods (page 136) for
overviews of these procedures.
2. For all chunks that migrate away from this shard, you do not need to do anything at this time. You do not
need to delete these documents from the shard because the chunks are automatically filtered out from queries by
mongos. You can remove these documents from the shard, if you like, at your leisure.
3. For chunks that migrate to this shard after the most recent backup, you must manually recover the chunks using
backups of other shards, or some other source. To determine what chunks have moved, view the changelog
collection in the Config Database (page 547).
Restore a Sharded Cluster

202

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Overview The procedure outlined in this document addresses how to restore an entire sharded cluster. For information on related backup procedures consider the following tutorials which describe backup procedures in greater
detail:
Backup a Sharded Cluster with Filesystem Snapshots (page 199)
Backup a Sharded Cluster with Database Dumps (page 200)
The exact procedure used to restore a database depends on the method used to capture the backup. See the MongoDB
Backup Methods (page 136) document for an overview of backups with MongoDB and Backup and Restore Sharded
Clusters (page 198) for a complete information on backups in MongoDB and backups of sharded clusters in particular.
Procedure
1. Stop all mongos and mongod processes, including all shards and all config servers.
2. If shard hostnames have changed, you must manually update the shards collection in the Config Database
(page 547) to use the new hostnames. Do the following:
(a) Start the three config servers (page 488) by issuing commands similar to the following, using values
appropriate to your configuration:
mongod --configsvr --dbpath /data/configdb --port 27019

(b) Restore the Config Database (page 547) on each config server.
(c) Start one mongos instance.
(d) Update the Config Database (page 547) collection named shards to reflect the new hostnames.
3. Restore the following:
Data files for each server in each shard. Because replica sets provide each production shard, restore all the
members of the replica set or use the other standard approaches for restoring a replica set from backup.
See the Restore a Snapshot (page 192) and Restore a Database with mongorestore (page 196) sections for
details on these procedures.
Data files for each config server (page 488), if you have not already done so in the previous step.
4. Restart all the mongos instances.
5. Restart all the shard mongod instances.
6. Restart all the config servers mongod instances.
7. Connect to a mongos instance from a mongo shell and use the db.printShardingStatus() method to
ensure that the cluster is operational, as follows:
db.printShardingStatus()
show collections

Recover Data after an Unexpected Shutdown

If MongoDB does not shutdown cleanly 79 the on-disk representation of the data files will likely reflect an inconsistent
state which could lead to data corruption. 80
79 To ensure a clean shut down, use the db.shutdownServer() from the mongo shell, your control script, the mongod --shutdown
option on Linux systems, Control-C when running mongod in interactive mode, or kill $(pidof mongod) or kill -2 $(pidof
mongod).
80 You can also use the db.collection.validate() method to test the integrity of a single collection. However, this process is time
consuming, and without journaling you can safely assume that the data is in an invalid state and you should either run the repair operation or resync
from an intact member of the replica set.

4.2. Administration Tutorials

203

MongoDB Documentation, Release 2.4.9

To prevent data inconsistency and corruption, always shut down the database cleanly and use the durability journaling.
MongoDB writes data to the journal, by default, every 100 milliseconds, such that MongoDB can always recover to a
consistent state even in the case of an unclean shutdown due to power loss or other system failure.
If you are not running as part of a replica set and do not have journaling enabled, use the following procedure to
recover data that may be in an inconsistent state. If you are running as part of a replica set, you should always restore
from a backup or restart the mongod instance with an empty dbpath and allow MongoDB to perform an initial sync
to restore the data.
See also:
The Administration (page 135) documents, including Replica Set Syncing (page 410), and the documentation on the
repair, repairpath, and journal settings.
Process

Indications When you are aware of a mongod instance running without journaling that stops unexpectedly and
youre not running with replication, you should always run the repair operation before starting MongoDB again. If
youre using replication, then restore from a backup and allow replication to perform an initial sync (page 410) to
restore data.
If the mongod.lock file in the data directory specified by dbpath, /data/db by default, is not a zero-byte file,
then mongod will refuse to start, and you will find a message that contains the following line in your MongoDB log
our output:
Unclean shutdown detected.

This indicates that you need to run mongod with the --repair option. If you run repair when the mongodb.lock
file exists in your dbpath, or the optional --repairpath, you will see a message that contains the following line:
old lock file: /data/db/mongod.lock. probably means unclean shutdown

If you see this message, as a last resort you may remove the lockfile and run the repair operation before starting the
database normally, as in the following procedure:

Overview

Warning: Recovering a member of a replica set.

Do not use this procedure to recover a member of a replica set. Instead you should either restore from a backup
(page 136) or perform an initial sync using data from an intact member of the set, as described in Resync a Member
of a Replica Set (page 449).

There are two processes to repair data files that result from an unexpected shutdown:
1. Use the --repair option in conjunction with the --repairpath option. mongod will read the existing
data files, and write the existing data to new data files. This does not modify or alter the existing data files.
You do not need to remove the mongod.lock file before using this procedure.
2. Use the --repair option. mongod will read the existing data files, write the existing data to new files and
replace the existing, possibly corrupt, files with new files.
You must remove the mongod.lock file before using this procedure.
Note: --repair functionality is also available in the shell with the db.repairDatabase() helper for the
repairDatabase command.

204

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Procedures To repair your data files using the --repairpath option to preserve the original data files unmodified:
1. Start mongod using --repair to read the existing data files.
mongod --dbpath /data/db --repair --repairpath /data/db0

When this completes, the new repaired data files will be in the /data/db0 directory.
2. Start mongod using the following invocation to point the dbpath at /data/db0:
mongod --dbpath /data/db0

Once you confirm that the data files are operational you may delete or archive the data files in the /data/db
directory.
To repair your data files without preserving the original files, do not use the --repairpath option, as in the
following procedure:
1. Remove the stale lock file:
rm /data/db/mongod.lock

Replace /data/db with your dbpath where your MongoDB instances data files reside.
Warning: After you remove the mongod.lock file you must run the --repair process before using
your database.
2. Start mongod using --repair to read the existing data files.
mongod --dbpath /data/db --repair

When this completes, the repaired data files will replace the original data files in the /data/db directory.
3. Start mongod using the following invocation to point the dbpath at /data/db:
mongod --dbpath /data/db

mongod.lock

In normal operation, you should never remove the mongod.lock file and start mongod. Instead consider the one
of the above methods to recover the database and remove the lock files. In dire situations you can remove the lockfile,
and start the database using the possibly corrupt files, and attempt to recover data from the database; however, its
impossible to predict the state of the database in these situations.
If you are not running with journaling, and your database shuts down unexpectedly for any reason, you should always
proceed as if your database is in an inconsistent and likely corrupt state. If at all possible restore from backup
(page 136) or, if running as a replica set, restore by performing an initial sync using data from an intact member of the
set, as described in Resync a Member of a Replica Set (page 449).

4.2.3 MongoDB Scripting

The mongo shell is an interactive JavaScript shell for MongoDB, and is part of all MongoDB distributions81 . This
section provides an introduction to the shell, and outlines key functions, operations, and use of the mongo shell. Also
consider FAQ: The mongo Shell (page 567) and the shell method and other relevant reference material.
Note: Most examples in the MongoDB Manual use the mongo shell; however, many drivers (page 95) provide
81 http://www.mongodb.org/downloads

4.2. Administration Tutorials

205

MongoDB Documentation, Release 2.4.9

similar interfaces to MongoDB.

Server-side JavaScript (page 206) Details MongoDBs support for executing JavaScript code for server-side operations.
Data Types in the mongo Shell (page 207) Describes the super-set of JSON available for use in the mongo shell.
Write Scripts for the mongo Shell (page 210) An introduction to the mongo shell for writing scripts to manipulate
data and administer MongoDB.
Getting Started with the mongo Shell (page 212) Introduces the use and operation of the MongoDB shell.
Access the mongo Shell Help Information (page 216) Describes the available methods for accessing online help for
the operation of the mongo interactive shell.
mongo Shell Quick Reference (page 217) A high level reference to the use and operation of the mongo shell.
Server-side JavaScript
Changed in version 2.4: The V8 JavaScript engine, which became the default in 2.4, allows multiple JavaScript
operations to execute at the same time. Prior to 2.4, MongoDB operations that required the JavaScript interpreter had
to acquire a lock, and a single mongod could only run a single JavaScript operation at a time.
Overview

MongoDB supports the execution of JavaScript code for the following server-side operations:
mapReduce and the corresponding mongo shell method db.collection.mapReduce(). See MapReduce (page 284) for more information.
eval command, and the corresponding mongo shell method db.eval()
$where operator
Running .js files via a mongo shell Instance on the Server (page 206)
JavaScript in MongoDB
Although the above operations use JavaScript, most interactions with MongoDB do not use JavaScript but use an
idiomatic driver (page 95) in the language of the interacting application.
See also:
Store a JavaScript Function on the Server (page 183)
You can disable all server-side execution of JavaScript, by passing the --noscripting option on the command
line or setting noscripting in a configuration file.
Running .js files via a mongo shell Instance on the Server

You can run a JavaScript (.js) file using a mongo shell instance on the server. This is a good technique for performing
batch administrative work. When you run mongo shell on the server, connecting via the localhost interface, the
connection is fast with low latency.
The command helpers (page 218) provided in the mongo shell are not available in JavaScript files because they are
not valid JavaScript. The following table maps the most common mongo shell helpers to their JavaScript equivalents.

206

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Shell Helpers
show dbs, show databases
use <db>
show collections
show users
show log <logname>
show logs
it

JavaScript Equivalents
db.adminCommand('listDatabases')
db = db.getSiblingDB('<db>')
db.getCollectionNames()
db.system.users.find()
db.adminCommand({ 'getLog' : '<logname>' })
db.adminCommand({ 'getLog' : '*' })
cursor = db.collection.find()
if ( cursor.hasNext() ){
cursor.next();
}

Concurrency

Refer to the individual method or operator documentation for any concurrency information. See also the concurrency
table (page 570).
Data Types in the mongo Shell
MongoDB BSON provide support for additional data types than JSON. Drivers (page 95) provide native support for
these data types in host languages and the mongo shell also provides several helper classes to support the use of these
data types in the mongo JavaScript shell. See MongoDB Extended JSON (page 227) for additional information.
Types

Date The mongo shell provides various options to return the date, either as a string or as an object:
Date() method which returns the current date as a string.
Date() constructor which returns an ISODate object when used with the new operator.
ISODate() constructor which returns an ISODate object when used with or without the new operator.
Consider the following examples:
To return the date as a string, use the Date() method, as in the following example:
var myDateString = Date();

To print the value of the variable, type the variable name in the shell, as in the following:
myDateString

The result is the value of myDateString:

4.2. Administration Tutorials

207

MongoDB Documentation, Release 2.4.9

Wed Dec 19 2012 01:03:25 GMT-0500 (EST)

To verify the type, use the typeof operator, as in the following:

typeof myDateString

The operation returns string.

To get the date as an ISODate object, instantiate a new instance using the Date() constructor with the new
operator, as in the following example:
var myDateObject = new Date();

To print the value of the variable, type the variable name in the shell, as in the following:
myDateObject

The result is the value of myDateObject:

ISODate("2012-12-19T06:01:17.171Z")

To verify the type, use the typeof operator, as in the following:

typeof myDateObject

The operation returns object.

To get the date as an ISODate object, instantiate a new instance using the ISODate() constructor without
the new operator, as in the following example:
var myDateObject2 = ISODate();

You can use the new operator with the ISODate() constructor as well.
To print the value of the variable, type the variable name in the shell, as in the following:
myDateObject2

The result is the value of myDateObject2:

ISODate("2012-12-19T06:15:33.035Z")

To verify the type, use the typeof operator, as in the following:

typeof myDateObject2

The operation returns object.

ObjectId The mongo shell provides the ObjectId() wrapper class around ObjectId data types. To generate a
new ObjectId, use the following operation in the mongo shell:
new ObjectId

See
ObjectId (page 129) for full documentation of ObjectIds in MongoDB.

208

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

NumberLong By default, the mongo shell treats all numbers as floating-point values. The mongo shell provides
the NumberLong() class to handle 64-bit integers.
The NumberLong() constructor accepts the long as a string:
NumberLong("2090845886852")

The following examples use the NumberLong() class to write to the collection:
db.collection.insert( { _id: 10, calc: NumberLong("2090845886852") } )
db.collection.update( { _id: 10 },
{ $set: { calc: NumberLong("2555555000000") } } )
db.collection.update( { _id: 10 },
{ $inc: { calc: NumberLong(5) } } )

Retrieve the document to verify:

db.collection.findOne( { _id: 10 } )

In the returned document, the calc field contains a NumberLong object:

{ "_id" : 10, "calc" : NumberLong("2555555000005") }

If you use the $inc to increment the value of a field that contains a NumberLong object by a float, the data type
changes to a floating point value, as in the following example:
1. Use $inc to increment the calc field by 5, which the mongo shell treats as a float:
db.collection.update( { _id: 10 },
{ $inc: { calc: 5 } } )

2. Retrieve the updated document:

db.collection.findOne( { _id: 10 } )

In the updated document, the calc field contains a floating point value:
{ "_id" : 10, "calc" : 2555555000010 }

NumberInt By default, the mongo shell treats all numbers as floating-point values. The mongo shell provides the
NumberInt() constructor to explicitly specify 32-bit integers.
Check Types in the mongo Shell

To determine the type of fields, the mongo shell provides the following operators:
instanceof returns a boolean to test if a value has a specific type.
typeof returns the type of a field.
Example
Consider the following operations using instanceof and typeof:
The following operation tests whether the _id field is of type ObjectId:
mydoc._id instanceof ObjectId

The operation returns true.

The following operation returns the type of the _id field:

4.2. Administration Tutorials

209

MongoDB Documentation, Release 2.4.9

typeof mydoc._id

In this case typeof will return the more generic object type rather than ObjectId type.

Write Scripts for the mongo Shell

You can write scripts for the mongo shell in JavaScript that manipulate data in MongoDB or perform administrative
operation. For more information about the mongo shell see MongoDB Scripting (page 205), and see the Running .js
files via a mongo shell Instance on the Server (page 206) section for more information about using these mongo script.
This tutorial provides an introduction to writing JavaScript that uses the mongo shell to access MongoDB.
Opening New Connections

From the mongo shell or from a JavaScript file, you can instantiate database connections using the Mongo() constructor:
new Mongo()
new Mongo(<host>)
new Mongo(<host:port>)

Consider the following example that instantiates a new connection to the MongoDB instance running on localhost on
the default port and sets the global db variable to myDatabase using the getDB() method:
conn = new Mongo();
db = conn.getDB("myDatabase");

Additionally, you can use the connect() method to connect to the MongoDB instance. The following example
connects to the MongoDB instance that is running on localhost with the non-default port 27020 and set the
global db variable:
db = connect("localhost:27020/myDatabase");

Differences Between Interactive and Scripted mongo

When writing scripts for the mongo shell, consider the following:
To set the db global variable, use the getDB() method or the connect() method. You can assign the
database reference to a variable other than db.
Inside the script, call db.getLastError() explicitly to wait for the result of write operations (page 42).
You cannot use any shell helper (e.g. use <dbname>, show dbs, etc.) inside the JavaScript file because
they are not valid JavaScript.
The following table maps the most common mongo shell helpers to their JavaScript equivalents.

210

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Shell Helpers
show dbs, show databases
use <db>
show collections
show users
show log <logname>
show logs
it

JavaScript Equivalents
db.adminCommand('listDatabases')
db = db.getSiblingDB('<db>')
db.getCollectionNames()
db.system.users.find()
db.adminCommand({ 'getLog' : '<logname>' })
db.adminCommand({ 'getLog' : '*' })
cursor = db.collection.find()
if ( cursor.hasNext() ){
cursor.next();
}

In interactive mode, mongo prints the results of operations including the content of all cursors. In scripts, either
use the JavaScript print() function or the mongo specific printjson() function which returns formatted
JSON.
Example
To print all items in a result cursor in mongo shell scripts, use the following idiom:
cursor = db.collection.find();
while ( cursor.hasNext() ) {
printjson( cursor.next() );
}

Scripting

From the system prompt, use mongo to evaluate JavaScript.

--eval option Use the --eval option to mongo to pass the shell a JavaScript fragment, as in the following:
mongo test --eval "printjson(db.getCollectionNames())"

This returns the output of db.getCollectionNames() using the mongo shell connected to the mongod or
mongos instance running on port 27017 on the localhost interface.
Execute a JavaScript file You can specify a .js file to the mongo shell, and mongo will execute the JavaScript
directly. Consider the following example:
mongo localhost:27017/test myjsfile.js

This operation executes the myjsfile.js script in a mongo shell that connects to the test database on the
mongod instance accessible via the localhost interface on port 27017.

4.2. Administration Tutorials

211

MongoDB Documentation, Release 2.4.9

Alternately, you can specify the mongodb connection parameters inside of the javascript file using the Mongo()
constructor. See Opening New Connections (page 210) for more information.
You can execute a .js file from within the mongo shell, using the load() function, as in the following:
load("myjstest.js")

This function loads and executes the myjstest.js file.

The load() method accepts relative and absolute paths. If the current working directory of the mongo shell is
/data/db, and the myjstest.js resides in the /data/db/scripts directory, then the following calls within
the mongo shell would be equivalent:
load("scripts/myjstest.js")
load("/data/db/scripts/myjstest.js")

Note: There is no search path for the load() function. If the desired script is not in the current working directory
or the full specified path, mongo will not be able to access the file.

Getting Started with the mongo Shell

This document provides a basic introduction to using the mongo shell. See Install MongoDB (page 3) for instructions
on installing MongoDB for your system.
Start the mongo Shell

To start the mongo shell and connect to your MongoDB instance running on localhost with default port:
1. Go to your <mongodb installation dir>:
cd <mongodb installation dir>

2. Type ./bin/mongo to start mongo:

./bin/mongo

If you have added the <mongodb installation dir>/bin to the PATH environment variable, you can
just type mongo instead of ./bin/mongo.
3. To display the database you are using, type db:
db

The operation should return test, which is the default database. To switch databases, issue the use <db>
helper, as in the following example:
use <database>

To list the available databases, use the helper show dbs. See also How can I access different databases
temporarily? (page 568) to access a different database from the current database without switching your current
database context (i.e. db..)
To start the mongo shell with other options, see examples of starting up mongo and mongo reference which
provides details on the available options.
Note: When starting, mongo checks the users HOME directory for a JavaScript file named .mongorc.js. If found,
mongo interprets the content of .mongorc.js before displaying the prompt for the first time. If you use the shell

212

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

to evaluate a JavaScript file or expression, either by using the --eval option on the command line or by specifying
a .js file to mongo, mongo will read the .mongorc.js file after the JavaScript has finished processing.

Executing Queries

From the mongo shell, you can use the shell methods to run queries, as in the following example:
db.<collection>.find()

The db refers to the current database.

The <collection> is the name of the collection to query. See Collection Help (page 216) to list the available
collections.
If the mongo shell does not accept the name of the collection, for instance if the name contains a space, hyphen,
or starts with a number, you can use an alternate syntax to refer to the collection, as in the following:
db["3test"].find()
db.getCollection("3test").find()

The find() method is the JavaScript method to retrieve documents from <collection>. The find()
method returns a cursor to the results; however, in the mongo shell, if the returned cursor is not assigned to a
variable using the var keyword, then the cursor is automatically iterated up to 20 times to print up to the first
20 documents that match the query. The mongo shell will prompt Type it to iterate another 20 times.
You can set the DBQuery.shellBatchSize attribute to change the number of iteration from the default
value 20, as in the following example which sets it to 10:
DBQuery.shellBatchSize = 10;

For more information and examples on cursor handling in the mongo shell, see Cursors (page 35).
See also Cursor Help (page 217) for list of cursor help in the mongo shell.
For more documentation of basic MongoDB operations in the mongo shell, see:
Getting Started with MongoDB (page 19)
mongo Shell Quick Reference (page 217)
Read Operations (page 31)
Write Operations (page 42)
Indexing Tutorials (page 339)
Print

The mongo shell automatically prints the results of the find() method if the returned cursor is not assigned to
a variable using the var keyword. To format the result, you can add the .pretty() to the operation, as in the
following:
db.<collection>.find().pretty()

In addition, you can use the following explicit print methods in the mongo shell:
print() to print without formatting
print(tojson(<obj>)) to print with JSON formatting and equivalent to printjson()
4.2. Administration Tutorials

213

MongoDB Documentation, Release 2.4.9

printjson() to print with JSON formatting and equivalent to print(tojson(<obj>))

Evaluate a JavaScript File

You can execute a .js file from within the mongo shell, using the load() function, as in the following:
load("myjstest.js")

This function loads and executes the myjstest.js file.

The load() method accepts relative and absolute paths. If the current working directory of the mongo shell is
/data/db, and the myjstest.js resides in the /data/db/scripts directory, then the following calls within
the mongo shell would be equivalent:
load("scripts/myjstest.js")
load("/data/db/scripts/myjstest.js")

Note: There is no search path for the load() function. If the desired script is not in the current working directory
or the full specified path, mongo will not be able to access the file.

Use a Custom Prompt

You may modify the content of the prompt by creating the variable prompt in the shell. The prompt variable can
hold strings as well as any arbitrary JavaScript. If prompt holds a function that returns a string, mongo can display
dynamic information in each prompt. Consider the following examples:
Example
Create a prompt with the number of operations issued in the current session, define the following variables:
cmdCount = 1;
prompt = function() {
return (cmdCount++) + "> ";
}

The prompt would then resemble the following:

1> db.collection.find()
2> show collections
3>

Example
To create a mongo shell prompt in the form of <database>@<hostname>$ define the following variables:
host = db.serverStatus().host;
prompt = function() {
return db+"@"+host+"$ ";
}

The prompt would then resemble the following:

<database>@<hostname>$ use records
switched to db records
records@<hostname>$

214

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Example
To create a mongo shell prompt that contains the system up time and the number of documents in the current database,
define the following prompt variable:
prompt = function() {
return "Uptime:"+db.serverStatus().uptime+" Documents:"+db.stats().objects+" > ";
}

The prompt would then resemble the following:

Uptime:5897 Documents:6 > db.people.save({name : "James"});
Uptime:5948 Documents:7 >

Use an External Editor in the mongo Shell

New in version 2.2.

In the mongo shell you can use the edit operation to edit a function or variable in an external editor. The edit
operation uses the value of your environments EDITOR variable.
At your system prompt you can define the EDITOR variable and start mongo with the following two operations:
export EDITOR=vim
mongo

Then, consider the following example shell session:

MongoDB shell version: 2.2.0
> function f() {}
> edit f
> f
function f() {
print("this really works");
}
> f()
this really works
> o = {}
{ }
> edit o
> o
{ "soDoes" : "this" }
>

Note: As mongo shell interprets code edited in an external editor, it may modify code in functions, depending on
the JavaScript compiler. For mongo may convert 1+1 to 2 or remove comments. The actual changes affect only the
appearance of the code and will vary based on the version of JavaScript used but will not affect the semantics of the
code.

Exit the Shell

To exit the shell, type quit() or use the <Ctrl-c> shortcut.

4.2. Administration Tutorials

215

MongoDB Documentation, Release 2.4.9

Access the mongo Shell Help Information

In addition to the documentation in the MongoDB Manual, the mongo shell provides some additional information
in its online help system. This document provides an overview of accessing this help information.
See also:
mongo Manual Page
MongoDB Scripting (page 205), and
mongo Shell Quick Reference (page 217).
Command Line Help

To see the list of options and help for starting the mongo shell, use the --help option from the command line:
mongo --help

Shell Help

To see the list of help, in the mongo shell, type help:

help

Database Help

To see the list of databases on the server, use the show dbs command:
show dbs

New in version 2.4: show databases is now an alias for show dbs
To see the list of help for methods you can use on the db object, call the db.help() method:
db.help()

To see the implementation of a method in the shell, type the db.<method name> without the parenthesis
(()), as in the following example which will return the implementation of the method db.addUser():
db.addUser

Collection Help

To see the list of collections in the current database, use the show collections command:
show collections

To see the help for methods available on the collection objects (e.g.
db.<collection>.help() method:

db.<collection>), use the

db.collection.help()

<collection> can be the name of a collection that exists, although you may specify a collection that doesnt
exist.

216

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

To see the collection method implementation, type the db.<collection>.<method> name without the
parenthesis (()), as in the following example which will return the implementation of the save() method:
db.collection.save

Cursor Help

When you perform read operations (page 31) with the find() method in the mongo shell, you can use various
cursor methods to modify the find() behavior and various JavaScript methods to handle the cursor returned from
the find() method.
To list the available modifier and cursor handling methods, use the db.collection.find().help()
command:
db.collection.find().help()

<collection> can be the name of a collection that exists, although you may specify a collection that doesnt
exist.
To see the implementation of the cursor method, type the db.<collection>.find().<method> name
without the parenthesis (()), as in the following example which will return the implementation of the
toArray() method:
db.collection.find().toArray

Some useful methods for handling cursors are:

hasNext() which checks whether the cursor has more documents to return.
next() which returns the next document and advances the cursor position forward by one.
forEach(<function>) which iterates the whole cursor and applies the <function> to each document
returned by the cursor. The <function> expects a single argument which corresponds to the document from
each iteration.
For examples on iterating a cursor and retrieving the documents from the cursor, see cursor handling (page 35). See
also js-query-cursor-methods for all available cursor methods.
Type Help

To get a list of the wrapper classes available in the mongo shell, such as BinData(), type help misc in the
mongo shell:
help misc

mongo Shell Quick Reference

mongo Shell Command History

You can retrieve previous commands issued in the mongo shell with the up and down arrow keys. Command history
is stored in ~/.dbshell file. See .dbshell for more information.

4.2. Administration Tutorials

217

MongoDB Documentation, Release 2.4.9

Command Line Options

The mongo executable can be started with numerous options. See mongo executable page for details on all
available options.
The following table displays some common options for mongo:
OpDescription
tion
--help Show command line options
--nodb Start mongo shell without connecting to a database.
To connect later, see Opening New Connections (page 210).
--shellUsed in conjunction with a JavaScript file (i.e. <file.js>) to continue in the mongo shell after running
the JavaScript file.
See JavaScript file (page 211) for an example.
Command Helpers

The mongo shell provides various help. The following table displays some common help methods and commands:
Help Methods and Description
Commands
help
Show help.
db.help()
Show help for database methods.
db.<collection>.help()
Show help on collection methods. The <collection> can be the name of an existing
collection or a non-existing collection.
show dbs
Print a list of all databases on the server.
use <db>
Switch current database to <db>. The mongo shell variable db is set to the current
database.
show
Print a list of all collections for current database
collections
show users
Print a list of users for current database.
show profile
Print the five most recent operations that took 1 millisecond or more. See documentation
on the database profiler (page 174) for more information.
show databases
New in version 2.4: Print a list of all available databases.
load()
Execute a JavaScript file. See Getting Started with the mongo Shell (page 212) for more
information.
Basic Shell JavaScript Operations

The mongo shell provides numerous http://docs.mongodb.org/manualreference/method methods

for database operations.
In the mongo shell, db is the variable that references the current database. The variable is automatically set to the
default database test or is set when you use the use <db> to switch current database.
The following table displays some common JavaScript operations:

218

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

JavaScript Database Operations

db.auth()
coll = db.<collection>

Description
If running in secure mode, authenticate the user.
Set a specific collection in the current database to a variable coll, as in the following example:
coll = db.myCollection;
You can perform operations on the myCollection
using the variable, as in the following example:
coll.find();

find()

Find all documents in the collection and returns a cursor.

See the db.collection.find() and Query Documents (page 60) for more information and examples.
See Cursors (page 35) for additional information on cursor handling in the mongo shell.
Insert a new document into the collection.
Update an existing document in the collection.
See Write Operations (page 42) for more information.
Insert either a new document or update an existing document in the collection.
See Write Operations (page 42) for more information.
Delete documents from the collection.
See Write Operations (page 42) for more information.
Drops or removes completely the collection.
Create a new index on the collection if the index does
not exist; otherwise, the operation has no effect.
Return a reference to another database using this same
connection without explicitly switching the current
database. This allows for cross database queries. See
How can I access different databases temporarily?
(page 568) for more information.

insert()
update()
save()

remove()
drop()
ensureIndex()
db.getSiblingDB()

For more information on performing operations in the shell, see:

MongoDB CRUD Concepts (page 29)
Read Operations (page 31)
Write Operations (page 42)
http://docs.mongodb.org/manualreference/method
Keyboard Shortcuts

Changed in version 2.2.

The mongo shell provides most keyboard shortcuts similar to those found in the bash shell or in Emacs. For some
functions mongo provides multiple key bindings, to accommodate several familiar paradigms.
The following table enumerates the keystrokes supported by the mongo shell:
Keystroke
Up-arrow
Down-arrow
Home
End
Tab

4.2. Administration Tutorials

Function
previous-history
next-history
beginning-of-line
end-of-line
autocomplete
Continued on next page
219

MongoDB Documentation, Release 2.4.9

Table 4.1 continued from previous page

Keystroke
Function
Left-arrow
backward-character
Right-arrow
forward-character
Ctrl-left-arrow
backward-word
Ctrl-right-arrow
forward-word
Meta-left-arrow
backward-word
Meta-right-arrow
forward-word
Ctrl-A
beginning-of-line
Ctrl-B
backward-char
Ctrl-C
exit-shell
Ctrl-D
delete-char (or exit shell)
Ctrl-E
end-of-line
Ctrl-F
forward-char
Ctrl-G
abort
Ctrl-J
accept-line
Ctrl-K
kill-line
Ctrl-L
clear-screen
Ctrl-M
accept-line
Ctrl-N
next-history
Ctrl-P
previous-history
Ctrl-R
reverse-search-history
Ctrl-S
forward-search-history
Ctrl-T
transpose-chars
Ctrl-U
unix-line-discard
Ctrl-W
unix-word-rubout
Ctrl-Y
yank
Ctrl-Z
Suspend (job control works in linux)
Ctrl-H (i.e. Backspace) backward-delete-char
Ctrl-I (i.e. Tab)
complete
Meta-B
backward-word
Meta-C
capitalize-word
Meta-D
kill-word
Meta-F
forward-word
Meta-L
downcase-word
Meta-U
upcase-word
Meta-Y
yank-pop
Meta-[Backspace]
backward-kill-word
Meta-<
beginning-of-history
Meta->
end-of-history

Queries

In the mongo shell, perform read operations using the find() and findOne() methods.
The find() method returns a cursor object which the mongo shell iterates to print documents on screen. By default,
mongo prints the first 20. The mongo shell will prompt the user to Type it to continue iterating the next 20
results.
The following table provides some common read operations in the mongo shell:

220

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Read Operations
db.collection.find(<query>)

db.collection.find( <query>,
<projection> )

db.collection.find().sort( <sort
order> )

db.collection.find( <query> ).sort(

<sort order> )
db.collection.find( ... ).limit( <n>
)
db.collection.find( ... ).skip( <n>
)
count()
db.collection.find( <query> ).count()

db.collection.findOne( <query> )

4.2. Administration Tutorials

Description
Find the documents matching the <query> criteria in
the collection. If the <query> criteria is not specified
or is empty (i.e {} ), the read operation selects all documents in the collection.
The following example selects the documents in the
users collection with the name field equal to "Joe":
coll = db.users;
coll.find( { name: "Joe" } );
For more information on specifying the <query> criteria, see Query Documents (page 60).
Find documents matching the <query> criteria and return just specific fields in the <projection>.
The following example selects all documents from the
collection but returns only the name field and the _id
field. The _id is always returned unless explicitly specified to not return.
coll = db.users;
coll.find( { },
{ name: true }
);
For
more
information
on
specifying
the
<projection>, see Limit Fields to Return from
a Query (page 64).
Return results in the specified <sort order>.
The following example selects all documents from the
collection and returns the results sorted by the name
field in ascending order (1). Use -1 for descending order:
coll = db.users;
coll.find().sort( { name: 1 } );
Return the documents matching the <query> criteria
in the specified <sort order>.
Limit result to <n> rows. Highly recommended if you
need only a certain number of rows for best performance.
Skip <n> results.
Returns total number of documents in the collection.
Returns the total number of documents that match the
query.
The count() ignores limit() and skip(). For
example, if 100 records match but the limit is 10,
count() will return 100. This will be faster than iterating yourself, but still take time.
Find and return a single document. Returns null if not
found.
The following example selects a single document in the users collection with the
name field matches to "Joe":
coll = db.users;
coll.findOne( { name: "Joe" } );
Internally, the findOne() method is the find()
method with a limit(1).
221

MongoDB Documentation, Release 2.4.9

See Query Documents (page 60) and Read Operations (page 31) documentation for more information and examples.
See http://docs.mongodb.org/manualreference/operator to specify other query operators.
Error Checking Methods

The mongo shell provides numerous administrative database methods, including error checking methods. These
methods are:
Error Checking Methods
db.getLastError()
db.getLastErrorObj()

Description
Returns error message from the last operation.
Returns the error document from the last operation.

Administrative Command Helpers

The following table lists some common methods to support database administration:
JavaScript Database
Description
Administration Methods
db.cloneDatabase(<host>)
Clone the current database from the <host> specified. The <host> database
instance must be in noauth mode.
db.copyDatabase(<from>,Copy the <from> database from the <host> to the <to> database on the
<to>, <host>)
current server.
The <host> database instance must be in noauth mode.
db.fromColl.renameCollection(<toColl>)
Rename collection from fromColl to <toColl>.
db.repairDatabase()
Repair and compact the current database. This operation can be very slow on
large databases.
db.addUser( <user>,
Add user to current database.
<pwd> )
db.getCollectionNames()Get the list of all collections in the current database.
db.dropDatabase()
Drops the current database.
See also administrative database methods for a full list of methods.
Opening Additional Connections

You can create new connections within the mongo shell.

The following table displays the methods to create the connections:
JavaScript Connection Create Methods
db = connect("<host><:port>/<dbname>")
conn = new Mongo()
db = conn.getDB("dbname")

Description
Open a new database connection.
Open a connection to a new server using new
Mongo().
Use getDB() method of the connection to select a
database.

See also Opening New Connections (page 210) for more information on the opening new connections from the mongo
shell.
Miscellaneous

The following table displays some miscellaneous methods:

222

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Method
Object.bsonsize(<document>)

Description
Prints the BSON size of a <document>

See the MongoDB JavaScript API Documentation82 for a full list of JavaScript methods .
Additional Resources

Consider the following reference material that addresses the mongo shell and its interface:
http://docs.mongodb.org/manualreference/program/mongo
http://docs.mongodb.org/manualreference/method
http://docs.mongodb.org/manualreference/operator
http://docs.mongodb.org/manualreference/command
Aggregation Reference (page 308)
Additionally, the MongoDB source code repository includes a jstests directory83 which contains numerous mongo
shell scripts.
See also:
The MongoDB Manual contains administrative documentation and tutorials though out several sections. See Replica
Set Tutorials (page 419) and Sharded Cluster Tutorials (page 506) for additional tutorials and information.

4.3 Administration Reference

UNIX ulimit Settings (page 223) Describes user resources limits (i.e. ulimit) and introduces the considerations
and optimal configurations for systems that run MongoDB deployments.
System Collections (page 227) Introduces the internal collections that MongoDB uses to track per-database metadata,
including indexes, collections, and authentication credentials.
MongoDB Extended JSON (page 227) Describes the JSON super set used to express BSON documents in the
mongo shell and other MongoDB tools.
Database Profiler Output (page 230) Describes the data collected by MongoDBs operation profiler, which introspects operations and reports data for analysis on performance and behavior.
Journaling Mechanics (page 234) Describes the internal operation of MongoDBs journaling facility and outlines
how the journal allows MongoDB to provide provides durability and crash resiliency.
Exit Codes and Statuses (page 235) Lists the unique codes returned by mongos and mongod processes upon exit.

4.3.1 UNIX ulimit Settings

Most UNIX-like operating systems, including Linux and OS X, provide ways to limit and control the usage of system
resources such as threads, files, and network connections on a per-process and per-user basis. These ulimits prevent
single users from using too many system resources. Sometimes, these limits have low default values that can cause a
number of issues in the course of normal MongoDB operation.
Note: Red Hat Enterprise Linux and CentOS 6 place a max process limitation of 1024 which overrides ulimit settings. Create a file named /etc/security/limits.d/99-mongodb-nproc.conf with new soft nproc
82 http://api.mongodb.org/js/index.html
83 https://github.com/mongodb/mongo/tree/master/jstests/

4.3. Administration Reference

223

MongoDB Documentation, Release 2.4.9

and hard nproc values to increase the process limit. See /etc/security/limits.d/90-nproc.conf file
as an example.

Resource Utilization
mongod and mongos each use threads and file descriptors to track connections and manage internal operations. This
section outlines the general resource utilization patterns for MongoDB. Use these figures in combination with the
actual information about your deployment and its use to determine ideal ulimit settings.
Generally, all mongod and mongos instances:
track each incoming connection with a file descriptor and a thread.
track each internal thread or pthread as a system process.
mongod

1 file descriptor for each data file in use by the mongod instance.
1 file descriptor for each journal file used by the mongod instance when journal is true.
In replica sets, each mongod maintains a connection to all other members of the set.
mongod uses background threads for a number of internal processes, including TTL collections (page 163), replication, and replica set health checks, which may require a small number of additional resources.
mongos

In addition to the threads and file descriptors for client connections, mongos must maintain connects to all config
servers and all shards, which includes all members of all replica sets.
For mongos, consider the following behaviors:
mongos instances maintain a connection pool to each shard so that the mongos can reuse connections and
quickly fulfill requests without needing to create new connections.
You can limit the number of incoming connections using the maxConns run-time option.
By restricting the number of incoming connections you can prevent a cascade effect where the mongos creates
too many connections on the mongod instances.
Note: You cannot set maxConns to a value higher than 20000.

Review and Set Resource Limits

ulimit

Note: Both the hard and the soft ulimit affect MongoDBs performance. The hard ulimit refers to the
maximum number of processes that a user can have active at any time. This is the ceiling: no non-root process can
increase the hard ulimit. In contrast, the soft ulimit is the limit that is actually enforced for a session or
process, but any process can increase it up to hard ulimit maximum.

224

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

A low soft ulimit can cause cant create new thread, closing connection errors if the number
of connections grows too high. For this reason, it is extremely important to set both ulimit values to the recommended values.
You can use the ulimit command at the system prompt to check system limits, as in the following example:
$ ulimit -a
-t: cpu time (seconds)
-f: file size (blocks)
-d: data seg size (kbytes)
-s: stack size (kbytes)
-c: core file size (blocks)
-m: resident set size (kbytes)
-u: processes
-n: file descriptors
-l: locked-in-memory size (kb)
-v: address space (kb)
-x: file locks
-i: pending signals
-q: bytes in POSIX msg queues
-e: max nice
-r: max rt priority
-N 15:

unlimited
unlimited
unlimited
8192
0
unlimited
192276
21000
40000
unlimited
unlimited
192276
819200
30
65
unlimited

ulimit refers to the per-user limitations for various resources. Therefore, if your mongod instance executes as
a user that is also running multiple processes, or multiple mongod processes, you might see contention for these
resources. Also, be aware that the processes value (i.e. -u) refers to the combined number of distinct processes
and sub-process threads.
You can change ulimit settings by issuing a command in the following form:
ulimit -n <value>

For many distributions of Linux you can change values by substituting the -n option for any possible value in the
output of ulimit -a. On OS X, use the launchctl limit command. See your operating system documentation
for the precise procedure for changing system limits on running systems.
Note: After changing the ulimit settings, you must restart the process to take advantage of the modified settings.
You can use the http://docs.mongodb.org/manualproc file system to see the current limitations on a
running process.
Depending on your systems configuration, and default settings, any change to system limits made using ulimit
may revert following system a system restart. Check your distribution and operating system documentation for more
information.

/proc File System

Note: This section applies only to Linux operating systems.

The http://docs.mongodb.org/manualproc file-system stores the per-process limits in the file system object located at /proc/<pid>/limits, where <pid> is the processs PID or process identifier. You can use the
following bash function to return the content of the limits object for a process or processes with a given name:
return-limits(){
for process in $@; do

4.3. Administration Reference

225

MongoDB Documentation, Release 2.4.9

process_pids=`ps -C $process -o pid --no-headers | cut -d " " -f 2`

if [ -z $@ ]; then
echo "[no $process running]"
else
for pid in $process_pids; do
echo "[$process #$pid -- limits]"
cat /proc/$pid/limits
done
fi
done
}

You can copy and paste this function into a current shell session or load it as part of a script. Call the function with
one the following invocations:
return-limits mongod
return-limits mongos
return-limits mongod mongos

The output of the first command may resemble the following:

[mongod #6809 -- limits]
Limit
Max cpu time
Max file size
Max data size
Max stack size
Max core file size
Max resident set
Max processes
Max open files
Max locked memory
Max address space
Max file locks
Max pending signals
Max msgqueue size
Max nice priority
Max realtime priority
Max realtime timeout

Soft Limit
unlimited
unlimited
unlimited
8720000
0
unlimited
192276
1024
40960000
unlimited
unlimited
192276
819200
30
65
unlimited

Hard Limit
unlimited
unlimited
unlimited
unlimited
unlimited
unlimited
192276
4096
40960000
unlimited
unlimited
192276
819200
30
65
unlimited

Units
seconds
bytes
bytes
bytes
bytes
bytes
processes
files
bytes
bytes
locks
signals
bytes

us

Recommended Settings
Every deployment may have unique requirements and settings; however, the following thresholds and settings are
particularly important for mongod and mongos deployments:
-f (file size): unlimited
-t (cpu time): unlimited
-v (virtual memory): unlimited 84
-n (open files): 64000
-m (memory size): unlimited 1
84 If you limit virtual or resident memory size on a system running MongoDB the operating system will refuse to honor additional allocation
requests.

226

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

-u (processes/threads): 32000
Always remember to restart your mongod and mongos instances after changing the ulimit settings to make sure
that the settings change takes effect.

4.3.2 System Collections

Synopsis
MongoDB stores system information in collections that use the <database>.system.* namespace, which MongoDB reserves for internal use. Do not create collections that begin with system.
MongoDB also stores some additional instance-local metadata in the local database (page 472), specifically for replication purposes.
Collections
System collections include these collections stored directly in the database:
<database>.system.namespaces
The <database>.system.namespaces (page 227) collection contains information about all of the
databases collections. Additional namespace metadata exists in the database.ns files and is opaque to
database users.
<database>.system.indexes
The <database>.system.indexes (page 227) collection lists all the indexes in the database. Add and
remove data from this collection via the ensureIndex() and dropIndex()
<database>.system.profile
The <database>.system.profile (page 227) collection stores database profiling information. For information on profiling, see Database Profiling (page 144).
<database>.system.users
The <database>.system.users (page 272) collection stores credentials for users who have access
to the database. For more information on this collection, see Add a User to a Database (page 259) and
<database>.system.users (page 272).
<database>.system.js
The <database>.system.js (page 227) collection holds special JavaScript code for use in server side
JavaScript (page 206). See Store a JavaScript Function on the Server (page 183) for more information.

4.3.3 MongoDB Extended JSON

MongoDB import and export utilities (page 150) (i.e. mongoimport and mongoexport) and MongoDB REST
Interfaces85 render an approximation of MongoDB BSON documents in JSON format.
The REST interface supports three different modes for document output:
Strict mode that produces output that conforms to the JSON RFC specifications86 .
JavaScript mode that produces output that most JavaScript interpreters can process (via the --jsonp option)
mongo Shell mode produces output that the mongo shell can process. This is extended JavaScript format.
85 http://docs.mongodb.org/ecosystem/tools/http-interfaces
86 http://www.json.org

4.3. Administration Reference

227

MongoDB Documentation, Release 2.4.9

MongoDB can process these representations in REST input.

Special representations of BSON data in JSON format make it possible to render information that have no obvious
corresponding JSON. In some cases MongoDB supports multiple equivalent representations of the same type information.
BSON Data Types and Associated Representations
The following presents the BSON data types and the associated representations in the three different modes.
Binary

data_binary
Strict Mode
{
"$binary": "<bindata>",
"$type": "<t>"
}

JavaScript Mode (via JSONP)

mongo Shell Mode

{
"$binary": "<bindata>",
"$type": "<t>"
}

BinData ( <t>, <bindata> )

<bindata> is the base64 representation of a binary string.

<t> is the hexadecimal representation of a single byte that indicates the data type.
Date

data_date
Strict Mode
{
"$date": <date>
}

JavaScript Mode (via JSONP)

mongo Shell Mode

new Date( <date> )

new Date ( <date> )

<date> is the JSON representation of a 64-bit signed integer for milliseconds since epoch UTC (unsigned
before version 1.9.1).
Timestamp

data_timestamp
Strict Mode
{
"$timestamp": {
"t": <t>,
"i": <i>
}
}

JavaScript Mode (via JSONP)

mongo Shell Mode

{
"$timestamp": {
"t": <t>,
"i": <i>
}
}

Timestamp( <t>, <i> )

<t> is the JSON representation of a 32-bit unsigned integer for seconds since epoch.
<i> is a 32-bit unsigned integer for the increment.
228

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Regular Expression

data_regex
Strict Mode

JavaScript Mode (via JSONP)

{
/<jRegex>/<jOptions>
"$regex": "<sRegex>",
"$options": "<sOptions>"
}

mongo Shell Mode

/<jRegex>/<jOptions>

<sRegex> is a string of valid JSON characters.

<jRegex> is a string that may contain valid JSON characters and unescaped double quote (") characters, but
may not contain unescaped forward slash (http://docs.mongodb.org/manual) characters.
<sOptions> is a string containing the regex options represented by the letters of the alphabet.
<jOptions> is a string that may contain only the characters g, i, m and s (added in v1.9). Because
the JavaScript and mongo Shell representations support a limited range of options, any nonconforming
options will be dropped when converting to this representation.
OID

data_oid
Strict Mode
{
"$oid": "<id>"
}

JavaScript Mode (via JSONP)

mongo Shell Mode

{
"$oid": "<id>"
}

ObjectId( "<id>" )

<id> is a 24-character hexadecimal string.

DB Reference

data_ref
Strict Mode
{
"$ref": "<name>",
"$id": "<id>"
}

JavaScript Mode (via JSONP)

mongo Shell Mode

{
"$ref" : "<name>",
"$id" : "<id>"
}

DBRef("<name>", "<id>")

<name> is a string of valid JSON characters.

<id> is any valid extended JSON type.

4.3. Administration Reference

229

MongoDB Documentation, Release 2.4.9

Undefined Type

data_undefined
Strict Mode
{
"$undefined": true
}

JavaScript Mode (via JSONP)

mongo Shell Mode

undefined

undefined

The representation for the JavaScript/BSON undefined type.

MinKey

data_minkey
Strict Mode
{
"$minKey": 1
}

JavaScript Mode (via JSONP)

mongo Shell Mode

{
"$minKey": 1
}

MinKey

The representation of the MinKey BSON data type that compares lower than all other types. See What is the
compare order for BSON types? (page 563) for more information on comparison order for BSON types.
MaxKey

data_maxkey
Strict Mode
{
"$maxKey": 1
}

JavaScript Mode (via JSONP)

mongo Shell Mode

{
"$maxKey": 1
}

MaxKey

The representation of the MaxKey BSON data type that compares higher than all other types. See What is the
compare order for BSON types? (page 563) for more information on comparison order for BSON types.

4.3.4 Database Profiler Output

The database profiler captures data information about read and write operations, cursor operations, and database commands. To configure the database profile and set the thresholds for capturing profile data, see the Analyze Performance
of Database Operations (page 174) section.
The database profiler writes data in the system.profile (page 227) collection, which is a capped collection. To
view the profilers output, use normal MongoDB queries on the system.profile (page 227) collection.
Note: Because the database profiler writes data to the system.profile (page 227) collection in a database, the
profiler will profile some write activity, even for databases that are otherwise read-only.

230

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

Example system.profile Document

The documents in the system.profile (page 227) collection have the following form. This example document
reflects an update operation:
{
"ts" : ISODate("2012-12-10T19:31:28.977Z"),
"op" : "update",
"ns" : "social.users",
"query" : {
"name" : "j.r."
},
"updateobj" : {
"$set" : {
"likes" : [
"basketball",
"trekking"
]
}
},
"nscanned" : 8,
"scanAndOrder" : true,
"moved" : true,
"nmoved" : 1,
"nupdated" : 1,
"keyUpdates" : 0,
"numYield" : 0,
"lockStats" : {
"timeLockedMicros" : {
"r" : NumberLong(0),
"w" : NumberLong(258)
},
"timeAcquiringMicros" : {
"r" : NumberLong(0),
"w" : NumberLong(7)
}
},
"millis" : 0,
"client" : "127.0.0.1",
"user" : ""
}

Output Reference
For any single operation, the documents created by the database profiler will include a subset of the following fields.
The precise selection of fields in these documents depends on the type of operation.
system.profile.ts
The timestamp of the operation.
system.profile.op
The type of operation. The possible values are:
insert
query
update

4.3. Administration Reference

231

MongoDB Documentation, Release 2.4.9

remove
getmore
command
system.profile.ns
The namespace the operation targets. Namespaces in MongoDB take the form of the database, followed by a
dot (.), followed by the name of the collection.
system.profile.query
The query document (page 60) used.
system.profile.command
The command operation.
system.profile.updateobj
The <update> document passed in during an update (page 42) operation.
system.profile.cursorid
The ID of the cursor accessed by a getmore operation.
system.profile.ntoreturn
Changed in version 2.2: In 2.0, MongoDB includes this field for query and command operations. In 2.2, this
information MongoDB also includes this field for getmore operations.
The number of documents the operation specified to return. For example, the profile command would
return one document (a results document) so the ntoreturn (page 232) value would be 1. The limit(5)
command would return five documents so the ntoreturn (page 232) value would be 5.
If the ntoreturn (page 232) value is 0, the command did not specify a number of documents to return, as
would be the case with a simple find() command with no limit specified.
system.profile.ntoskip
New in version 2.2.
The number of documents the skip() method specified to skip.
system.profile.nscanned
The number of documents that MongoDB scans in the index (page 313) in order to carry out the operation.
In general, if nscanned (page 232) is much higher than nreturned (page 233), the database is scanning
many objects to find the target objects. Consider creating an index to improve this.
system.profile.scanAndOrder
scanAndOrder (page 232) is a boolean that is true when a query cannot use the order of documents in the
index for returning sorted results: MongoDB must sort the documents after it receives the documents from a
cursor.
If scanAndOrder (page 232) is false, MongoDB can use the order of the documents in an index to return
sorted results.
system.profile.moved
This field appears with a value of true when an update operation moved one or more documents to a new
location on disk. If the operation did not result in a move, this field does not appear. Operations that result in a
move take more time than in-place updates and typically occur as a result of document growth.
system.profile.nmoved
New in version 2.2.
The number of documents the operation moved on disk. This field appears only if the operation resulted in a
move. The fields implicit value is zero, and the field is present only when non-zero.

232

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

system.profile.nupdated
New in version 2.2.
The number of documents updated by the operation.
system.profile.keyUpdates
New in version 2.2.
The number of index (page 313) keys the update changed in the operation. Changing an index key carries a
small performance cost because the database must remove the old key and inserts a new key into the B-tree
index.
system.profile.numYield
New in version 2.2.
The number of times the operation yielded to allow other operations to complete. Typically, operations yield
when they need access to data that MongoDB has not yet fully read into memory. This allows other operations
that have data in memory to complete while MongoDB reads in data for the yielding operation. For more
information, see the FAQ on when operations yield (page 570).
system.profile.lockStats
New in version 2.2.
The time in microseconds the operation spent acquiring and holding locks. This field reports data for the
following lock types:
R - global read lock
W - global write lock
r - database-specific read lock
w - database-specific write lock
system.profile.lockStats.timeLockedMicros
The time in microseconds the operation held a specific lock. For operations that require more than one
lock, like those that lock the local database to update the oplog, this value may be longer than the total
length of the operation (i.e. millis (page 233).)
system.profile.lockStats.timeAcquiringMicros
The time in microseconds the operation spent waiting to acquire a specific lock.
system.profile.nreturned
The number of documents returned by the operation.
system.profile.responseLength
The length in bytes of the operations result document. A large responseLength (page 233) can affect
performance. To limit the size of the result document for a query operation, you can use any of the following:
Projections (page 64)
The limit() method
The batchSize() method
Note: When MongoDB writes query profile information to the log, the responseLength (page 233) value
is in a field named reslen.
system.profile.millis
The time in milliseconds from the perspective of the mongod from the beginning of the operation to the end of
the operation.

4.3. Administration Reference

233

MongoDB Documentation, Release 2.4.9

system.profile.client
The IP address or hostname of the client connection where the operation originates.
For some operations, such as db.eval(), the client is 0.0.0.0:0 instead of an actual client.
system.profile.user
The authenticated user who ran the operation.

4.3.5 Journaling Mechanics

When running with journaling, MongoDB stores and applies write operations (page 42) in memory and in the on-disk
journal before the changes are present in the data files on disk. This document discusses the implementation and
mechanics of journaling in MongoDB systems. See Manage Journaling (page 181) for information on configuring,
tuning, and managing journaling.
Journal Files
With journaling enabled, MongoDB creates a journal subdirectory within the directory defined by dbpath, which is
/data/db by default. The journal directory holds journal files, which contain write-ahead redo logs. The directory
also holds a last-sequence-number file. A clean shutdown removes all the files in the journal directory. A dirty shutdown (crash) leaves files in the journal directory; these are used to automatically recover the database to a consistent
state when the mongod process is restarted.
Journal files are append-only files and have file names prefixed with j._. When a journal file holds 1 gigabyte of data,
MongoDB creates a new journal file. Once MongoDB applies all the write operations in a particular journal file to the
database data files, it deletes the file, as it is no longer needed for recovery purposes. Unless you write many bytes of
data per second, the journal directory should contain only two or three journal files.
You can use the smallfiles run time option when starting mongod to limit the size of each journal file to 128
megabytes, if you prefer.
To speed the frequent sequential writes that occur to the current journal file, you can ensure that the journal directory
is on a different filesystem from the database data files.
Important: If you place the journal on a different filesystem from your data files you cannot use a filesystem snapshot
alone to capture valid backups of a dbpath directory. In this case, use fsyncLock() to ensure that database files
are consistent before the snapshot and fsyncUnlock() once the snapshot is complete.
Note: Depending on your filesystem, you might experience a preallocation lag the first time you start a mongod
instance with journaling enabled.
MongoDB may preallocate journal files if the mongod process determines that it is more efficient to preallocate
journal files than create new journal files as needed. The amount of time required to pre-allocate lag might last several
minutes, during which you will not be able to connect to the database. This is a one-time preallocation and does not
occur with future invocations.
To avoid preallocation lag, see Avoid Preallocation Lag (page 182).
Storage Views used in Journaling
Journaling adds three internal storage views to MongoDB.
The shared view stores modified data for upload to the MongoDB data files. The shared view is the only view
with direct access to the MongoDB data files. When running with journaling, mongod asks the operating system to

234

Chapter 4. Administration

MongoDB Documentation, Release 2.4.9

map your existing on-disk data files to the shared view virtual memory view. The operating system maps the files
but does not load them. MongoDB later loads data files into the shared view as needed.
The private view stores data for use with read operations (page 31). The private view is the first place
MongoDB applies new write operations (page 42). Upon a journal commit, MongoDB copies the changes made in
the private view to the shared view, where they are then available for uploading to the database data files.
The journal is an on-disk view that stores new write operations after MongoDB applies the operation to the private
view but before applying them to the data files. The journal provides durability. If the mongod instance were to
crash without having applied the writes to the data files, the journal could replay the writes to the shared view for
eventual upload to the data files.
How Journaling Records Write Operations
MongoDB copies the write operations to the journal in batches called group commits. These group commits help
minimize the performance impact of journaling, since a group commit must block all writers during the commit. See
journalCommitInterval for information on the default commit interval.
Journaling stores raw operations that allow MongoDB to reconstruct the following:
document insertion/updates
index modifications
metadata changes to the namespace files
creation and dropping of databases and their associated data files
As write operations (page 42) occur, MongoDB writes the data to the private view in RAM and then copies the
write operations in batches to the journal. The journal stores the operations on disk to ensure durability. Each journal
entry describes the bytes the write operation changed in the data files.
MongoDB next applies the journals write operations to the shared view. At this point, the shared view
becomes inconsistent with the data files.
At default intervals of 60 seconds, MongoDB asks the operating system to flush the shared view to disk. This
brings the data files up-to-date with the latest write operations. The operating system may choose to flush the shared
view to disk at a higher frequency than 60 seconds, particularly if the system is low on free memory.
When MongoDB flushes write operations to the data files, MongoDB notes which journal writes have been flushed.
Once a journal file contains only flushed writes, it is no longer needed for recovery, and MongoDB either deletes it or
recycles it for a new journal file.
As part of journaling, MongoDB routinely asks the operating system to remap the shared view to the private
view, in order to save physical RAM. Upon a new remapping, the operating system knows that physical memory
pages can be shared between the shared view and the private view mappings.
Note: The interaction between the shared view and the on-disk data files is similar to how MongoDB works
without journaling, which is that MongoDB asks the operating system to flush in-memory changes back to the data
files every 60 seconds.

4.3.6 Exit Codes and Statuses

MongoDB will return one of the following codes and statuses when exiting. Use this guide to interpret logs and when
troubleshooting issues with mongod and mongos instances.
0
Returned by MongoDB applications upon successful exit.

4.3. Administration Reference

235

MongoDB Documentation, Release 2.4.9

2
The specified options are in error or are incompatible with other options.
3
Returned by mongod if there is a mismatch between hostnames specified on the command line and in the
local.sources (page 474) collection. mongod may also return this status if oplog collection in the local
database is not readable.
4
The version of the database is different from the version supported by the mongod (or mongod.exe) instance.
The instance exits cleanly. Restart mongod with the --upgrade option to upgrade the database to the version
supported by this mongod instance.
5
Returned by mongod if a moveChunk operation fails to confirm a commit.
12
Returned by the mongod.exe process on Windows when it receives a Control-C, Close, Break or Shutdown
event.
14
Returned by MongoDB applications which encounter an unrecoverable error, an uncaught exception or uncaught
signal. The system exits without performing a clean shut down.
20
Message: ERROR: wsastartup failed <reason>
Returned by MongoDB applications on Windows following an error in the WSAStartup function.
Message: NT Service Error
Returned by MongoDB applications for Windows due to failures installing, starting or removing the NT Service
for the application.
45
Returned when a MongoDB application cannot open a file or cannot obtain a lock on a file.
47
MongoDB applications exit cleanly following a large clock skew (32768 milliseconds) event.
48
mongod exits cleanly if the server socket closes. The server socket is on port 27017 by default, or as specified
to the --port run-time option.

49
Returned by mongod.exe or mongos.exe on Windows when either receives a shutdown message from the
Windows Service Control Manager.
100
Returned by mongod when the process throws an uncaught exception.

236

Chapter 4. Administration

CHAPTER 5

Security

This section outlines basic security and risk management strategies and access control. The included tutorials outline
specific tasks for configuring firewalls, authentication, and system privileges.
Security Introduction (page 237) A high-level introduction to security and MongoDB deployments.
Security Concepts (page 239) The core documentation of security.
Access Control (page 239) Control access to MongoDB instances using authentication and authorization.
Network Exposure and Security (page 242) Discusses potential security risks related to the network and strategies for decreasing possible network-based attack vectors for MongoDB.
Security and MongoDB API Interfaces (page 244) Discusses potential risks related
JavaScript, HTTP and REST interfaces, including strategies to control those risks.

to

MongoDBs

Sharded Cluster Security (page 241) MongoDB controls access to sharded clusters with key files.
Security Tutorials (page 245) Tutorials for enabling and configuring security features for MongoDB.
Create a Vulnerability Report (page 265) Report a vulnerability in MongoDB.
Network Security Tutorials (page 245) Ensure that the underlying network configuration supports a secure operating environment for MongoDB deployments, and appropriately limits access to MongoDB deployments.
Access Control Tutorials (page 257) MongoDBs access control system provides role-based access control for
limiting access to MongoDB deployments. These tutorials describe procedures relevant for the operation
and maintenance of this access control system.
Security Reference (page 267) Reference for security related functions.

5.1 Security Introduction

As with all software running in a networked environment, administrators of MongoDB must consider security and risk
exposures for a MongoDB deployment. There are no magic solutions for risk mitigation, and maintaining a secure
MongoDB deployment is an ongoing process.

5.1.1 Defense in Depth

The documents in this section takes a Defense in Depth approach to securing MongoDB deployments and addresses a
number of different methods for managing risk and reducing risk exposure.

237

MongoDB Documentation, Release 2.4.9

The intent of a Defense In Depth approach is to ensure there are no exploitable points of failure in your deployment
that could allow an intruder or un-trusted party to access the data stored in the MongoDB database. The easiest and
most effective way to reduce the risk of exploitation is to run MongoDB in a trusted environment, limit access, follow
a system of least privilege, and follow best development and deployment practices.

5.1.2 Trusted Environments

The most effective way to reduce risk for MongoDB deployments is to run your entire MongoDB deployment, including all MongoDB components (i.e. mongod, mongos and application instances) in a trusted environment. Trusted
environments use the following strategies to control access:
Use network filter (e.g. firewall) rules that block all connections from unknown systems to MongoDB components.
Bind mongod and mongos instances to specific IP addresses to limit accessibility.
Limit MongoDB programs to non-public local networks, and virtual private networks.

5.1.3 Operational Practices to Reduce Risk

You may further reduce risk by controlling access (page 239) to the database by employing authentication and authorization. Require authentication (page 239) for access to MongoDB instances and require strong, complex, single purpose authentication credentials. This should be part of your internal security policy. Employ authorization (page 240)
and deploy a model of least privilege, where all users have only the amount of access they need to accomplish required
tasks and no more. See Access Control (page 239) for more information.
Follow the best application development and deployment practices, which includes: validating all inputs, managing
sessions, and application-level access control.
Always run the mongod or mongos process as a unique user with the minimum required permissions and access.
Never run a MongoDB program as a root or administrative users. The system users that run the MongoDB processes
should have robust authentication credentials that prevent unauthorized or casual access.
To further limit the environment, you can run the mongod or mongos process in a chroot environment. Both userbased access restrictions and chroot configuration follow recommended conventions for administering all daemon
processes on Unix-like systems.

5.1.4 Data Encryption

To support audit requirements, you may need to encrypt data stored in MongoDB. For best results, you can encrypt
this data in the application layer by encrypting the content of fields that hold secure data.
Additionally, MongoDB1 has a partnership2 with Gazzang3 to encrypt and secure sensitive data within MongoDB.
The solution encrypts data in real time and Gazzang provides advanced key management that ensures only authorized
processes can access this data. The Gazzang software ensures that the cryptographic keys remain safe and ensures
compliance with standards including HIPAA, PCI-DSS, and FERPA.
For more information, refer to the following resources: Datasheet4 and Webinar5 .
1 http://mongodb.com
2 https://www.mongodb.com/partners/technology/gazzang
3 http://www.gazzang.com/
4 http://www.gazzang.com/images/datasheet-zNcrypt-for-MongoDB.pdf
5 http://gazzang.com/resources/videos/partner-videos/item/209-gazzang-zncrypt-on-mongodb

238

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

5.1.5 Additional Security Strategies

MongoDB provides various strategies to reduce network risk, such as configuring MongoDB or configuring firewalls
for MongoDB. See Network Exposure and Security (page 242) for more information.
In addition, consider the strategies listed in Security and MongoDB API Interfaces (page 244) to reduce interfacerelated risks for the mongo shell, HTTP status interface and the REST API.
MongoDB Enterprise supports authentication using Kerberos. See Deploy MongoDB with Kerberos Authentication
(page 261).

5.1.6 Vulnerability Notification

MongoDB takes security very seriously. If you discover a vulnerability in MongoDB, or would like to know more
about our vulnerability reporting and response process, see the Create a Vulnerability Report (page 265) document.

5.2 Security Concepts

These documents introduce and address concepts and strategies related to security practices in MongoDB deployments.
Access Control (page 239) Control access to MongoDB instances using authentication and authorization.
Inter-Process Authentication (page 240) Components of a MongoDB sharded cluster or replica set deployment must
be able to authenticate to each other to perform routine internal operations.
Sharded Cluster Security (page 241) MongoDB controls access to sharded clusters with key files.
Network Exposure and Security (page 242) Discusses potential security risks related to the network and strategies
for decreasing possible network-based attack vectors for MongoDB.
Security and MongoDB API Interfaces (page 244) Discusses potential risks related to MongoDBs JavaScript,
HTTP and REST interfaces, including strategies to control those risks.

5.2.1 Access Control

MongoDB provides support for authentication and authorization on a per-database level. Users exist in the context of
a single logical database.
Authentication
MongoDB provisions authentication, or verification of the user identity, on a per-database level. Authentication disables anonymous access to the database. For basic authentication, MongoDB stores the user credentials in a databases
system.users (page 272) collection.
Authentication is disabled by default. To enable authentication for a given mongod or mongos instance, use the
auth and keyFile configuration settings. For details, see Enable Authentication (page 257).
For MongoDB Enterprise installations, authentication using a Kerberos service is available. See Deploy MongoDB
with Kerberos Authentication (page 261).
Important: You can authenticate as only one user for a given database. If you authenticate to a database as one user
and later authenticate on the same database as a different user, the second authentication invalidates the first. You can,
however, log into a different database as a different user and not invalidate your authentication on other databases,
though this is not a recommended approach.

5.2. Security Concepts

239

MongoDB Documentation, Release 2.4.9

Each client connection should authenticate as exactly one user.

Authorization
MongoDB provisions authorization, or access to databases and operations, on a per-database level. MongoDB uses
a role-based approach to authorization, storing each users roles in a privilege document (page 267) in a databases
system.users (page 272) collection. For more information on privilege documents and available user roles, see
system.users Privilege Documents (page 272) and User Privilege Roles in MongoDB (page 267).
Important: The admin database provides roles that are unavailable in other databases, including a role that effectively makes a user a MongoDB system superuser. See Database Administration Roles (page 269) and Administrative
Roles (page 269).
To assign roles to users, you must be a user with administrative role in the database. As such, you must first create an
administrative user. For details, see Create a User Administrator (page 258) and Add a User to a Database (page 259).
system.users Collection
A databases system.users (page 272) collection stores information for authentication and authorization to that
database. Specifically, the collection stores user credentials for authentication and user privilege information for
authorization. MongoDB requires authorization to access the system.users (page 272) collection in order to
prevent privilege escalation attacks. To access the collection, you must have either userAdmin (page 269) or
userAdminAnyDatabase (page 271) role.
Changed in version 2.4: The schema of system.users (page 272) changed to accommodate a more sophisticated
authorization using user privilege model, as defined in privilege documents (page 267).

5.2.2 Inter-Process Authentication

In most cases, replica set and sharded cluster administrators do not have to keep additional considerations in mind
beyond the normal security precautions that all MongoDB administrators must take. However, ensure that:
Your network configuration will allow every member of the replica set to contact every other member of the
replica set.
If you use MongoDBs authentication system to limit access to your infrastructure, ensure that you configure a
keyFile on all members to permit authentication.
For most instances, the most effective ways to control access and to secure the connection between members of a
replica set depend on network-level access control. Use your environments firewall and network routing to ensure
that traffic only from clients and other replica set members can reach your mongod instances. If needed, use virtual
private networks (VPNs) to ensure secure connections over wide area networks (WANs.)
Enable Authentication in Replica Sets and Sharded Clusters
New in version 1.8: Added support authentication in replica set deployments.
Changed in version 1.9.1: Added support authentication in sharded replica set deployments.
MongoDB provides an authentication mechanism for mongod and mongos instances connecting to replica sets.
These instances enable authentication but specify a shared key file that serves as a shared password.
To enable authentication, add the following option to your configuration file:

240

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

keyFile = /srv/mongodb/keyfile

Note: You may chose to set these run-time configuration options using the --keyFile (or mongos --keyFile)
options on the command line.
Setting keyFile enables authentication and specifies a key file for the replica set members to use when authenticating
to each other. The content of the key file is arbitrary but must be the same on all members of the replica set and on all
mongos instances that connect to the set.
The key file must be between 6 and 1024 characters and may only contain characters in the base64 set. The key file
must not have group or world permissions on UNIX systems. See Generate a Key File (page 261) for instructions
on generating a key file.

5.2.3 Sharded Cluster Security

In most respects security for sharded clusters similar to other MongoDB deployments. Sharded clusters use the
same keyfile (page 240) and access control (page 239) as all MongoDB deployments. However, there are additional
considerations when using authentication with sharded clusters.
Important: In addition to the mechanisms described in this section, always run sharded clusters in a trusted networking environment. Ensure that the network only permits trusted traffic to reach mongos and mongod instances.
See also:
Enable Authentication in a Sharded Cluster (page 513).
Access Control Privileges in Sharded Clusters
In sharded clusters, MongoDB provides separate administrative privileges for the sharded cluster and for each shard.
Sharded Cluster Authentication. When connected to a mongos, you can grant access to the sharded clusters
admin database. 6 These credentials reside on the config servers.
Users can access to the cluster according to their permissions (page 267). To receive privileges for the cluster,
you must authenticate while connected to a mongos instance.
Shard Server Authentication. To allow administrators to connect and authenticate directly to specific shards, create
users in the admin database on the mongod instance, or replica set, that provide each shard.
These users only have access to a single shard and are completely distinct from the cluster-wide credentials.
Important: Always connect and authenticate to sharded clusters via a mongos instance.
Beyond these proprieties, privileges for sharded clusters are functionally the same as any other MongoDB deployment.
See Access Control (page 239) for more information.
Access a Sharded Cluster with Authentication
To access a sharded cluster as an authenticated user, from the command line, use the authentication options when
connecting to a mongos. Or, you can connect first and then authenticate with the authenticate command or the
db.auth() method.
6 Credentials for databases other than the admin database reside in the mongod instance (or replica set) that is the primary shard (page 487)
for that database.

5.2. Security Concepts

241

MongoDB Documentation, Release 2.4.9

To close an authenticated session, see the logout command.

Restriction on localhost Interface
Sharded clusters have restrictions on the use of localhost interface. If the host identifier for a MongoDB instance is either localhost or 127.0.0.1, then you must use localhost or 127.0.0.1 to identify all
MongoDB instances in a deployment.This applies to the host argument to the addShard command as well as to
the --configdb option for the mongos. If you mix localhost addresses with remote host address, sharded clusters
will not function correctly.

5.2.4 Network Exposure and Security

By default, MongoDB programs (i.e. mongos and mongod) will bind to all available network interfaces (i.e. IP
addresses) on a system.
This page outlines various runtime options that allow you to limit access to MongoDB programs.
Configuration Options
You can limit the network exposure with the following mongod and and mongos configuration options:
nohttpinterface, rest, bind_ip, and port. You can use a configuration file to specify these
settings.
nohttpinterface

The nohttpinterface setting for mongod and mongos instances disables the home status page, which would
run on port 28017 by default. The status interface is read-only by default. You may also specify this option on the
command line as mongod --nohttpinterface or mongos --nohttpinterface.
Authentication does not control or affect access to this interface.
Important: Disable this option for production deployments. If you do leave this interface enabled, you should only
allow trusted clients to access this port. See Firewalls (page 243).

rest

The rest setting for mongod enables a fully interactive administrative REST interface, which is disabled by default.
The status interface, which is enabled by default, is read-only. This configuration makes that interface fully interactive.
The REST interface does not support any authentication and you should always restrict access to this interface to only
allow trusted clients to connect to this port.
You may also enable this interface on the command line as mongod --rest.
Important: Disable this option for production deployments. If do you leave this interface enabled, you should only
allow trusted clients to access this port.

242

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

bind_ip

The bind_ip setting for mongod and mongos instances limits the network interfaces on which MongoDB programs
will listen for incoming connections. You can also specify a number of interfaces by passing bind_ip a comma
separated list of IP addresses. You can use the mongod --bind_ip and mongos --bind_ip option on the
command line at run time to limit the network accessibility of a MongoDB program.
Important: Make sure that your mongod and mongos instances are only accessible on trusted networks. If your
system has more than one network interface, bind MongoDB programs to the private or internal network interface.

port

The port setting for mongod and mongos instances changes the main port on which the mongod or mongos
instance listens for connections. The default port is 27017. Changing the port does not meaningfully reduce risk or
limit exposure. You may also specify this option on the command line as mongod --port or mongos --port.
Setting port also indirectly sets the port for the HTTP status interface, which is always available on the port numbered
1000 greater than the primary mongod port.
Only allow trusted clients to connect to the port for the mongod and mongos instances. See Firewalls (page 243).
See also Security Considerations (page 147) and Default MongoDB Port (page 275).
Firewalls
Firewalls allow administrators to filter and control access to a system by providing granular control over what network
communications. For administrators of MongoDB, the following capabilities are important: limiting incoming traffic
on a specific port to specific systems, and limiting incoming traffic from untrusted hosts.
On Linux systems, the iptables interface provides access to the underlying netfilter firewall. On Windows
systems, netsh command line interface provides access to the underlying Windows Firewall. For additional information about firewall configuration, see Configure Linux iptables Firewall for MongoDB (page 245) and Configure
Windows netsh Firewall for MongoDB (page 249).
For best results and to minimize overall exposure, ensure that only traffic from trusted sources can reach mongod and
mongos instances and that the mongod and mongos instances can only connect to trusted outputs.
See also:
For MongoDB deployments on Amazons web services, see the Amazon EC27 page, which addresses Amazons
Security Groups and other EC2-specific security features.
Virtual Private Networks
Virtual private networks, or VPNs, make it possible to link two networks over an encrypted and limited-access trusted
network. Typically MongoDB users who use VPNs use SSL rather than IPSEC VPNs for performance issues.
Depending on configuration and implementation, VPNs provide for certificate validation and a choice of encryption
protocols, which requires a rigorous level of authentication and identification of all clients. Furthermore, because
VPNs provide a secure tunnel, by using a VPN connection to control access to your MongoDB instance, you can
prevent tampering and man-in-the-middle attacks.
7 http://docs.mongodb.org/ecosystem/platforms/amazon-ec2

5.2. Security Concepts

243

MongoDB Documentation, Release 2.4.9

5.2.5 Security and MongoDB API Interfaces

The following section contains strategies to limit risks related to MongoDBs available interfaces including JavaScript,
HTTP, and REST interfaces.
JavaScript and the Security of the mongo Shell
The following JavaScript evaluation behaviors of the mongo shell represents risk exposures.
JavaScript Expression or JavaScript File

The mongo program can evaluate JavaScript expressions using the command line --eval option. Also, the mongo
program can evaluate a JavaScript file (.js) passed directly to it (e.g. mongo someFile.js).
Because the mongo program evaluates the JavaScript directly, inputs should only come from trusted sources.
.mongorc.js File

If a .mongorc.js file exists 8 , the mongo shell will evaluate a .mongorc.js file before starting. You can disable
this behavior by passing the mongo --norc option.
HTTP Status Interface
The HTTP status interface provides a web-based interface that includes a variety of operational data, logs, and status
reports regarding the mongod or mongos instance. The HTTP interface is always available on the port numbered
1000 greater than the primary mongod port. By default, the HTTP interface port is 28017, but is indirectly set using
the port option which allows you to configure the primary mongod port.
Without the rest setting, this interface is entirely read-only, and limited in scope; nevertheless, this interface
may represent an exposure. To disable the HTTP interface, set the nohttpinterface run time option or the
--nohttpinterface command line option. See also Configuration Options (page 242).
REST API
The REST API to MongoDB provides additional information and write access on top of the HTTP Status interface.
While the REST API does not provide any support for insert, update, or remove operations, it does provide administrative access, and its accessibility represents a vulnerability in a secure environment. The REST interface is disabled
by default, and is not recommended for production use.
If you must use the REST API, please control and limit access to the REST API. The REST API does not include any
support for authentication, even when running with auth enabled.
See the following documents for instructions on restricting access to the REST API interface:
Configure Linux iptables Firewall for MongoDB (page 245)
Configure Windows netsh Firewall for MongoDB (page 249)
8 On Linux and Unix systems, mongo reads the .mongorc.js file from $HOME/.mongorc.js (i.e. ~/.mongorc.js). On Windows,
mongo.exe reads the .mongorc.js file from %HOME%.mongorc.js or %HOMEDRIVE%%HOMEPATH%.mongorc.js.

244

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

5.3 Security Tutorials

The following tutorials provide instructions for enabling and using the security features available in MongoDB.
Network Security Tutorials (page 245) Ensure that the underlying network configuration supports a secure operating
environment for MongoDB deployments, and appropriately limits access to MongoDB deployments.
Configure Linux iptables Firewall for MongoDB (page 245) Basic firewall configuration patterns and examples for iptables on Linux systems.
Configure Windows netsh Firewall for MongoDB (page 249) Basic firewall configuration patterns and examples for netsh on Windows systems.
Connect to MongoDB with SSL (page 252) SSL allows MongoDB clients to support encrypted connections to
mongod instances.
Access Control Tutorials (page 257) MongoDBs access control system provides role-based access control for limiting access to MongoDB deployments. These tutorials describe procedures relevant for the operation and
maintenance of this access control system.
Enable Authentication (page 257) Describes the process for enabling authentication for MongoDB deployments.
Create a User Administrator (page 258) Create users with special permissions to to create, modify, and remove
other users, as well as administer authentication credentials (e.g. passwords).
Add a User to a Database (page 259) Create non-administrator users using MongoDBs role-based authentication system.
Deploy MongoDB with Kerberos Authentication (page 261) Describes the process, for MongoDB Enterprise,
used to enable and implement a Kerberos-based authentication system for MongoDB deployments.
Create a Vulnerability Report (page 265) Report a vulnerability in MongoDB.

5.3.1 Network Security Tutorials

The following tutorials provide information on handling network security for MongoDB.
Configure Linux iptables Firewall for MongoDB (page 245) Basic firewall configuration patterns and examples for
iptables on Linux systems.
Configure Windows netsh Firewall for MongoDB (page 249) Basic firewall configuration patterns and examples for
netsh on Windows systems.
Connect to MongoDB with SSL (page 252) SSL allows MongoDB clients to support encrypted connections to
mongod instances.
Configure Linux iptables Firewall for MongoDB
On contemporary Linux systems, the iptables program provides methods for managing the Linux Kernels
netfilter or network packet filtering capabilities. These firewall rules make it possible for administrators to
control what hosts can connect to the system, and limit risk exposure by limiting the hosts that can connect to a
system.
This document outlines basic firewall configurations for iptables firewalls on Linux. Use these approaches as a
starting point for your larger networking organization. For a detailed overview of security practices and risk management for MongoDB, see Security Concepts (page 239).
See also:

5.3. Security Tutorials

245

MongoDB Documentation, Release 2.4.9

For MongoDB deployments on Amazons web services, see the Amazon EC29 page, which addresses Amazons
Security Groups and other EC2-specific security features.
Overview

Rules in iptables configurations fall into chains, which describe the process for filtering and processing specific
streams of traffic. Chains have an order, and packets must pass through earlier rules in a chain to reach later rules.
This document addresses only the following two chains:
INPUT Controls all incoming traffic.
OUTPUT Controls all outgoing traffic.
Given the default ports (page 242) of all MongoDB processes, you must configure networking rules that permit only
required communication between your application and the appropriate mongod and mongos instances.
Be aware that, by default, the default policy of iptables is to allow all connections and traffic unless explicitly
disabled. The configuration changes outlined in this document will create rules that explicitly allow traffic from
specific addresses and on specific ports, using a default policy that drops all traffic that is not explicitly allowed. When
you have properly configured your iptables rules to allow only the traffic that you want to permit, you can Change
Default Policy to DROP (page 248).
Patterns

This section contains a number of patterns and examples for configuring iptables for use with MongoDB deployments. If you have configured different ports using the port configuration setting, you will need to modify the rules
accordingly.
Traffic to and from mongod Instances This pattern is applicable to all mongod instances running as standalone
instances or as part of a replica set.
The goal of this pattern is to explicitly allow traffic to the mongod instance from the application server. In the
following examples, replace <ip-address> with the IP address of the application server:

iptables -A INPUT -s <ip-address> -p tcp --destination-port 27017 -m state --state NEW,ESTABLISHED -j

iptables -A OUTPUT -d <ip-address> -p tcp --source-port 27017 -m state --state ESTABLISHED -j ACCEPT

The first rule allows all incoming traffic from <ip-address> on port 27017, which allows the application server to
connect to the mongod instance. The second rule, allows outgoing traffic from the mongod to reach the application
server.
Optional
If you have only one application server, you can replace <ip-address> with either the IP address itself, such as:
198.51.100.55. You can also express this using CIDR notation as 198.51.100.55/32. If you want to permit
a larger block of possible IP addresses you can allow traffic from a http://docs.mongodb.org/manual24
using one of the following specifications for the <ip-address>, as follows:
10.10.10.10/24
10.10.10.10/255.255.255.0
9 http://docs.mongodb.org/ecosystem/platforms/amazon-ec2

246

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

Traffic to and from mongos Instances mongos instances provide query routing for sharded clusters. Clients
connect to mongos instances, which behave from the clients perspective as mongod instances. In turn, the mongos
connects to all mongod instances that are components of the sharded cluster.
Use the same iptables command to allow traffic to and from these instances as you would from the mongod
instances that are members of the replica set. Take the configuration outlined in the Traffic to and from mongod
Instances (page 246) section as an example.
Traffic to and from a MongoDB Config Server Config servers, host the config database that stores metadata
for sharded clusters. Each production cluster has three config servers, initiated using the mongod --configsvr
option. 10 Config servers listen for connections on port 27019. As a result, add the following iptables rules to the
config server to allow incoming and outgoing connection on port 27019, for connection to the other config servers.

iptables -A INPUT -s <ip-address> -p tcp --destination-port 27019 -m state --state NEW,ESTABLISHED -j

iptables -A OUTPUT -d <ip-address> -p tcp --source-port 27019 -m state --state ESTABLISHED -j ACCEPT

Replace <ip-address> with the address or address space of all the mongod that provide config servers.
Additionally, config servers need to allow incoming connections from all of the mongos instances in the cluster and
all mongod instances in the cluster. Add rules that resemble the following:

iptables -A INPUT -s <ip-address> -p tcp --destination-port 27019 -m state --state NEW,ESTABLISHED -j

Replace <ip-address> with the address of the mongos instances and the shard mongod instances.
Traffic to and from a MongoDB Shard Server For shard servers, running as mongod --shardsvr 11 Because
the default port number when running with shardsvr is 27018, you must configure the following iptables rules
to allow traffic to and from each shard:

iptables -A INPUT -s <ip-address> -p tcp --destination-port 27018 -m state --state NEW,ESTABLISHED -j

iptables -A OUTPUT -d <ip-address> -p tcp --source-port 27018 -m state --state ESTABLISHED -j ACCEPT

Replace the <ip-address> specification with the IP address of all mongod. This allows you to permit incoming
and outgoing traffic between all shards including constituent replica set members, to:
all mongod instances in the shards replica sets.
all mongod instances in other shards.

12

Furthermore, shards need to be able make outgoing connections to:

all mongos instances.
all mongod instances in the config servers.
Create a rule that resembles the following, and replace the <ip-address> with the address of the config servers
and the mongos instances:
iptables -A OUTPUT -d <ip-address> -p tcp --source-port 27018 -m state --state ESTABLISHED -j ACCEPT

Provide Access For Monitoring Systems

1. The mongostat diagnostic tool, when running with the --discover needs to be able to reach all components of a cluster, including the config servers, the shard servers, and the mongos instances.
10

You can also run a config server by setting the configsvr option in a configuration file.
You can also specify the shard server option using the shardsvr setting in the configuration file. Shard members are also often conventional
replica sets using the default port.
12 All shards in a cluster need to be able to communicate with all other shards to facilitate chunk and balancing operations.
11

5.3. Security Tutorials

247

MongoDB Documentation, Release 2.4.9

2. If your monitoring system needs access the HTTP interface, insert the following rule to the chain:

iptables -A INPUT -s <ip-address> -p tcp --destination-port 28017 -m state --state NEW,ESTABLISH

Replace <ip-address> with the address of the instance that needs access to the HTTP or REST interface.
For all deployments, you should restrict access to this port to only the monitoring instance.
Optional
For shard server mongod instances running with shardsvr, the rule would resemble the following:

iptables -A INPUT -s <ip-address> -p tcp --destination-port 28018 -m state --state NEW,ESTABLISH

For config server mongod instances running with configsvr, the rule would resemble the following:

iptables -A INPUT -s <ip-address> -p tcp --destination-port 28019 -m state --state NEW,ESTABLISH

Change Default Policy to DROP

The default policy for iptables chains is to allow all traffic. After completing all iptables configuration changes,
you must change the default policy to DROP so that all traffic that isnt explicitly allowed as above will not be able to
reach components of the MongoDB deployment. Issue the following commands to change this policy:
iptables -P INPUT DROP
iptables -P OUTPUT DROP

Manage and Maintain iptables Configuration

This section contains a number of basic operations for managing and using iptables. There are various front end
tools that automate some aspects of iptables configuration, but at the core all iptables front ends provide the
same basic functionality:
Make all iptables Rules Persistent By default all iptables rules are only stored in memory. When your
system restarts, your firewall rules will revert to their defaults. When you have tested a rule set and have guaranteed
that it effectively controls traffic you can use the following operations to you should make the rule set persistent.
On Red Hat Enterprise Linux, Fedora Linux, and related distributions you can issue the following command:
service iptables save

On Debian, Ubuntu, and related distributions, you can use the following command to dump the iptables rules to
the /etc/iptables.conf file:
iptables-save > /etc/iptables.conf

Run the following operation to restore the network rules:

iptables-restore < /etc/iptables.conf

Place this command in your rc.local file, or in the /etc/network/if-up.d/iptables file with other
similar operations.

248

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

List all iptables Rules To list all of currently applied iptables rules, use the following operation at the system
shell.
iptables --L

Flush all iptables Rules If you make a configuration mistake when entering iptables rules or simply need to
revert to the default rule set, you can use the following operation at the system shell to flush all rules:
iptables --F

If youve already made your iptables rules persistent, you will need to repeat the appropriate procedure in the
Make all iptables Rules Persistent (page 248) section.
Configure Windows netsh Firewall for MongoDB
On Windows Server systems, the netsh program provides methods for managing the Windows Firewall. These
firewall rules make it possible for administrators to control what hosts can connect to the system, and limit risk
exposure by limiting the hosts that can connect to a system.
This document outlines basic Windows Firewall configurations. Use these approaches as a starting point for your
larger networking organization. For a detailed over view of security practices and risk management for MongoDB, see
Security Concepts (page 239).
See also:
Windows Firewall13 documentation from Microsoft.
Overview

Windows Firewall processes rules in an ordered determined by rule type, and parsed in the following order:
1. Windows Service Hardening
2. Connection security rules
3. Authenticated Bypass Rules
4. Block Rules
5. Allow Rules
6. Default Rules
By default, the policy in Windows Firewall allows all outbound connections and blocks all incoming connections.
Given the default ports (page 242) of all MongoDB processes, you must configure networking rules that permit only
required communication between your application and the appropriate mongod.exe and mongos.exe instances.
The configuration changes outlined in this document will create rules which explicitly allow traffic from specific
addresses and on specific ports, using a default policy that drops all traffic that is not explicitly allowed.
You can configure the Windows Firewall with using the netsh command line tool or through a windows application.
On Windows Server 2008 this application is Windows Firewall With Advanced Security in Administrative Tools. On
previous versions of Windows Server, access the Windows Firewall application in the System and Security control
panel.
The procedures in this document use the netsh command line tool.
13 http://technet.microsoft.com/en-us/network/bb545423.aspx

5.3. Security Tutorials

249

MongoDB Documentation, Release 2.4.9

Patterns

This section contains a number of patterns and examples for configuring Windows Firewall for use with MongoDB
deployments. If you have configured different ports using the port configuration setting, you will need to modify the
rules accordingly.
Traffic to and from mongod.exe Instances This pattern is applicable to all mongod.exe instances running as
standalone instances or as part of a replica set. The goal of this pattern is to explicitly allow traffic to the mongod.exe
instance from the application server.

netsh advfirewall firewall add rule name="Open mongod port 27017" dir=in action=allow protocol=TCP lo

This rule allows all incoming traffic to port 27017, which allows the application server to connect to the
mongod.exe instance.
Windows Firewall also allows enabling network access for an entire application rather than to a specific port, as in the
following example:

netsh advfirewall firewall add rule name="Allowing mongod" dir=in action=allow program=" C:\mongodb\b

You can allow all access for a mongos.exe server, with the following invocation:

netsh advfirewall firewall add rule name="Allowing mongos" dir=in action=allow program=" C:\mongodb\b

Traffic to and from mongos.exe Instances mongos.exe instances provide query routing for sharded clusters.
Clients connect to mongos.exe instances, which behave from the clients perspective as mongod.exe instances.
In turn, the mongos.exe connects to all mongod.exe instances that are components of the sharded cluster.
Use the same Windows Firewall command to allow traffic to and from these instances as you would from the
mongod.exe instances that are members of the replica set.

netsh advfirewall firewall add rule name="Open mongod shard port 27018" dir=in action=allow protocol=

Traffic to and from a MongoDB Config Server Configuration servers, host the config database that stores metadata for sharded clusters. Each production cluster has three configuration servers, initiated using the mongod
--configsvr option. 14 Configuration servers listen for connections on port 27019. As a result, add the following Windows Firewall rules to the config server to allow incoming and outgoing connection on port 27019, for
connection to the other config servers.

netsh advfirewall firewall add rule name="Open mongod config svr port 27019" dir=in action=allow prot

Additionally, config servers need to allow incoming connections from all of the mongos.exe instances in the cluster
and all mongod.exe instances in the cluster. Add rules that resemble the following:

netsh advfirewall firewall add rule name="Open mongod config svr inbound" dir=in action=allow protoco

Replace <ip-address> with the addresses of the mongos.exe instances and the shard mongod.exe instances.
Traffic to and from a MongoDB Shard Server For shard servers, running as mongod --shardsvr 15 Because
the default port number when running with shardsvr is 27018, you must configure the following Windows Firewall
rules to allow traffic to and from each shard:
14

You can also run a config server by setting the configsvr option in a configuration file.
You can also specify the shard server option using the shardsvr setting in the configuration file. Shard members are also often conventional
replica sets using the default port.
15

250

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

netsh advfirewall firewall add rule name="Open mongod shardsvr inbound" dir=in action=allow protocol=
netsh advfirewall firewall add rule name="Open mongod shardsvr outbound" dir=out action=allow protoco

Replace the <ip-address> specification with the IP address of all mongod.exe instances. This allows you to
permit incoming and outgoing traffic between all shards including constituent replica set members to:
all mongod.exe instances in the shards replica sets.
all mongod.exe instances in other shards.

16

Furthermore, shards need to be able make outgoing connections to:

all mongos.exe instances.
all mongod.exe instances in the config servers.
Create a rule that resembles the following, and replace the <ip-address> with the address of the config servers
and the mongos.exe instances:

netsh advfirewall firewall add rule name="Open mongod config svr outbound" dir=out action=allow proto

Provide Access For Monitoring Systems

1. The mongostat diagnostic tool, when running with the --discover needs to be able to reach all components of a cluster, including the config servers, the shard servers, and the mongos.exe instances.
2. If your monitoring system needs access the HTTP interface, insert the following rule to the chain:

netsh advfirewall firewall add rule name="Open mongod HTTP monitoring inbound" dir=in action=all

Replace <ip-address> with the address of the instance that needs access to the HTTP or REST interface.
For all deployments, you should restrict access to this port to only the monitoring instance.
Optional
For shard server mongod.exe instances running with shardsvr, the rule would resemble the following:

netsh advfirewall firewall add rule name="Open mongos HTTP monitoring inbound" dir=in action=all

For config server mongod.exe instances running with configsvr, the rule would resemble the following:
netsh advfirewall firewall add rule name="Open mongod configsvr HTTP monitoring inbound" dir=in

Manage and Maintain Windows Firewall Configurations

This section contains a number of basic operations for managing and using netsh. While you can use the GUI front
ends to manage the Windows Firewall, all core functionality is accessible is accessible from netsh.
Delete all Windows Firewall Rules To delete the firewall rule allowing mongod.exe traffic:
netsh advfirewall firewall delete rule name="Open mongod port 27017" protocol=tcp localport=27017

netsh advfirewall firewall delete rule name="Open mongod shard port 27018" protocol=tcp localport=270
16

All shards in a cluster need to be able to communicate with all other shards to facilitate chunk and balancing operations.

5.3. Security Tutorials

251

MongoDB Documentation, Release 2.4.9

List All Windows Firewall Rules To return a list of all Windows Firewall rules:
netsh advfirewall firewall show rule name=all

Reset Windows Firewall

To reset the Windows Firewall rules:

netsh advfirewall reset

Backup and Restore Windows Firewall Rules To simplify administration of larger collection of systems, you can
export or import firewall systems from different servers) rules very easily on Windows:
Export all firewall rules with the following command:
netsh advfirewall export "C:\temp\MongoDBfw.wfw"

Replace "C:\temp\MongoDBfw.wfw" with a path of your choosing. You can use a command in the following
form to import a file created using this operation:
netsh advfirewall import "C:\temp\MongoDBfw.wfw"

Connect to MongoDB with SSL

This document outlines the use and operation of MongoDBs SSL support. SSL allows MongoDB clients to support
encrypted connections to mongod instances.
Note: The default distribution of MongoDB17 does not contain support for SSL. To use SSL, you must either build
MongoDB locally passing the --ssl option to scons or use MongoDB Enterprise18 .
These instructions outline the process for getting started with SSL and assume that you have already installed a build
of MongoDB that includes SSL support and that your client driver supports SSL.
Configure mongod and mongos for SSL

Combine SSL Certificate and Key File Before you can use SSL, you must have a .pem file that contains the public
key certificate and private key. MongoDB can use any valid SSL certificate. To generate a self-signed certificate and
private key, use a command that resembles the following:
cd /etc/ssl/
openssl req -new -x509 -days 365 -nodes -out mongodb-cert.crt -keyout mongodb-cert.key

This operation generates a new, self-signed certificate with no passphrase that is valid for 365 days. Once you have
the certificate, concatenate the certificate and private key to a .pem file, as in the following example:
cat mongodb-cert.key mongodb-cert.crt > mongodb.pem

Set Up mongod and mongos with SSL Certificate and Key To use SSL in your MongoDB deployment, include
the following run-time options with mongod and mongos:
sslOnNormalPorts
sslPEMKeyFile with the .pem file that contains the SSL certificate and key.
17 http://www.mongodb.org/downloads
18 http://www.mongodb.com/products/mongodb-enterprise

252

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

Consider the following syntax for mongod:

mongod --sslOnNormalPorts --sslPEMKeyFile <pem>

For example, given an SSL certificate located at /etc/ssl/mongodb.pem, configure mongod to use SSL encryption for all connections with the following command:
mongod --sslOnNormalPorts --sslPEMKeyFile /etc/ssl/mongodb.pem

Note:
Specify <pem> with the full path name to the certificate.
If the private key portion of the <pem> is encrypted, specify the encryption password with the
sslPEMKeyPassword option.
You may also specify these options in the configuration file, as in the following example:
sslOnNormalPorts = true
sslPEMKeyFile = /etc/ssl/mongodb.pem

To connect, to mongod and mongos instances using SSL, the mongo shell and MongoDB tools must include the
--ssl option. See SSL Configuration for Clients (page 254) for more information on connecting to mongod and
mongos running with SSL.
Set Up mongod and mongos with Certificate Validation To set up mongod or mongos for SSL encryption
using an SSL certificate signed by a certificate authority, include the following run-time options during startup:
sslOnNormalPorts
sslPEMKeyFile with the name of the .pem file that contains the signed SSL certificate and key.
sslCAFile with the name of the .pem file that contains the root certificate chain from the Certificate Authority.
Consider the following syntax for mongod:
mongod --sslOnNormalPorts --sslPEMKeyFile <pem> --sslCAFile <ca>

For example, given a signed SSL certificate located at /etc/ssl/mongodb.pem and the certificate authority file
at /etc/ssl/ca.pem, you can configure mongod for SSL encryption as follows:
mongod --sslOnNormalPorts --sslPEMKeyFile /etc/ssl/mongodb.pem --sslCAFile /etc/ssl/ca.pem

Note:
Specify the <pem> file and the <ca> file with either the full path name or the relative path name.
If the <pem> is encrypted, specify the encryption password with the sslPEMKeyPassword option.
You may also specify these options in the configuration file, as in the following example:
sslOnNormalPorts = true
sslPEMKeyFile = /etc/ssl/mongodb.pem
sslCAFile = /etc/ssl/ca.pem

To connect, to mongod and mongos instances using SSL, the mongo tools must include the both the --ssl and
--sslPEMKeyFile option. See SSL Configuration for Clients (page 254) for more information on connecting to
mongod and mongos running with SSL.

5.3. Security Tutorials

253

MongoDB Documentation, Release 2.4.9

Block Revoked Certificates for Clients To prevent clients with revoked certificates from connecting, include the
sslCRLFile to specify a .pem file that contains revoked certificates.
For example, the following mongod with SSL configuration includes the sslCRLFile setting:

mongod --sslOnNormalPorts --sslCRLFile /etc/ssl/ca-crl.pem --sslPEMKeyFile /etc/ssl/mongodb.pem --ssl

Clients with revoked certificates in the /etc/ssl/ca-crl.pem will not be able to connect to this mongod instance.
Validate Only if a Client Presents a Certificate In most cases it is important to ensure that clients present valid
certificates. However, if you have clients that cannot present a client certificate, or are transitioning to using a certificate
authority you may only want to validate certificates from clients that present a certificate.
If you want to bypass validation for clients that dont present certificates, include the
sslWeakCertificateValidation run-time option with mongod and mongos. If the client does not
present a certificate, no validation occurs. These connections, though not validated, are still encrypted using SSL.
For example, consider the following mongod
sslWeakCertificateValidation setting:

with

an

SSL

configuration

that

includes

the

mongod --sslOnNormalPorts --sslWeakCertificateValidation --sslPEMKeyFile /etc/ssl/mongodb.pem --sslCA

Then, clients can connect either with the option --ssl and no certificate or with the option --ssl and a valid
certificate. See SSL Configuration for Clients (page 254) for more information on SSL connections for clients.
Note: If the client presents a certificate, the certificate must be a valid certificate.
All connections, including those that have not presented certificates are encrypted using SSL.

Run in FIPS Mode If your mongod or mongos is running on a system with an OpenSSL library configured with
the FIPS 140-2 module, you can run mongod or mongos in FIPS mode, with the sslFIPSMode setting.
SSL Configuration for Clients

Clients must have support for SSL to work with a mongod or a mongos instance that has SSL support enabled. The
current versions of the Python, Java, Ruby, Node.js, .NET, and C++ drivers have support for SSL, with full support
coming in future releases of other drivers.
mongo SSL Configuration For SSL connections, you must use the mongo shell built with SSL support or distributed with MongoDB Enterprise. To support SSL, mongo has the following settings:
--ssl
--sslPEMKeyFile with the name of the .pem file that contains the SSL certificate and key.
--sslCAFile with the name of the .pem file that contains the certificate from the Certificate Authority.
--sslPEMKeyPassword option if the client certificate-key file is encrypted.
Connect to MongoDB Instance with SSL Encryption To connect to a mongod or mongos instance that requires
only a SSL encryption mode (page 252), start mongo shell with --ssl, as in the following:
mongo --ssl

254

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

Connect to MongoDB Instance that Requires Client Certificates To connect to a mongod or mongos that requires CA-signed client certificates (page 253), start the mongo shell with --ssl and the --sslPEMKeyFile
option to specify the signed certificate-key file, as in the following:
mongo --ssl --sslPEMKeyFile /etc/ssl/client.pem

Connect to MongoDB Instance that Validates when Presented with a Certificate To connect to a mongod or
mongos instance that only requires valid certificates when the client presents a certificate (page 254), start mongo
shell either with the --ssl ssl and no certificate or with the --ssl ssl and a valid signed certificate.
For example, if mongod is running with weak certificate validation, both of the following mongo shell clients can
connect to that mongod:
mongo --ssl
mongo --ssl --sslPEMKeyFile /etc/ssl/client.pem

Important: If the client presents a certificate, the certificate must be valid.

MMS Monitoring Agent The Monitoring agent will also have to connect via SSL in order to gather its stats. Because the agent already utilizes SSL for its communications to the MMS servers, this is just a matter of enabling SSL
support in MMS itself on a per host basis.
Use the Edit host button (i.e. the pencil) on the Hosts page in the MMS console to enable SSL.
Please see the MMS documentation19 for more information about MMS configuration.
PyMongo Add the ssl=True parameter to a PyMongo MongoClient20 to create a MongoDB connection to
an SSL MongoDB instance:
from pymongo import MongoClient
c = MongoClient(host="mongodb.example.net", port=27017, ssl=True)

To connect to a replica set, use the following operation:

from pymongo import MongoReplicaSetClient
c = MongoReplicaSetClient("mongodb.example.net:27017",
replicaSet="mysetname", ssl=True)

PyMongo also supports an ssl=true option for the MongoDB URI:

mongodb://mongodb.example.net:27017/?ssl=true

Java Consider the following example SSLApp.java class file:

import com.mongodb.*;
import javax.net.ssl.SSLSocketFactory;
public class SSLApp {
public static void main(String args[])

throws Exception {

MongoClientOptions o = new MongoClientOptions.Builder()

19 http://mms.mongodb.com/help
20 http://api.mongodb.org/python/current/api/pymongo/mongo_client.html#pymongo.mongo_client.MongoClient

5.3. Security Tutorials

255

MongoDB Documentation, Release 2.4.9

.socketFactory(SSLSocketFactory.getDefault())
.build();
MongoClient m = new MongoClient("localhost", o);
DB db = m.getDB( "test" );
DBCollection c = db.getCollection( "foo" );
System.out.println( c.findOne() );
}
}

Ruby The recent versions of the Ruby driver have support for connections to SSL servers. Install the latest version
of the driver with the following command:
gem install mongo

Then connect to a standalone instance, using the following form:

require 'rubygems'
require 'mongo'
connection = MongoClient.new('localhost', 27017, :ssl => true)

Replace connection with the following if youre connecting to a replica set:

connection = MongoReplicaSetClient.new(['localhost:27017'],
['localhost:27018'],
:ssl => true)

Here, mongod instance run on localhost:27017 and localhost:27018.

Node.JS (node-mongodb-native) In the node-mongodb-native21 driver, use the following invocation to connect to a mongod or mongos instance via SSL:
var db1 = new Db(MONGODB, new Server("127.0.0.1", 27017,
{ auto_reconnect: false, poolSize:4, ssl:ssl } );

To connect to a replica set via SSL, use the following form:

var replSet = new ReplSetServers( [
new Server( RS.host, RS.ports[1], { auto_reconnect: true } ),
new Server( RS.host, RS.ports[0], { auto_reconnect: true } ),
],
{rs_name:RS.name, ssl:ssl}
);

.NET As of release 1.6, the .NET driver supports SSL connections with mongod and mongos instances. To connect
using SSL, you must add an option to the connection string, specifying ssl=true as follows:
var connectionString = "mongodb://localhost/?ssl=true";
var server = MongoServer.Create(connectionString);

The .NET driver will validate the certificate against the local trusted certificate store, in addition to providing encryption of the server. This behavior may produce issues during testing if the server uses a self-signed certificate. If
21 https://github.com/mongodb/node-mongodb-native

256

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

you encounter this issue, add the sslverifycertificate=false option to the connection string to prevent the
.NET driver from validating the certificate, as follows:
var connectionString = "mongodb://localhost/?ssl=true&sslverifycertificate=false";
var server = MongoServer.Create(connectionString);

5.3.2 Access Control Tutorials

The following tutorials provide instructions on how to enable authentication and limit access for users with privilege
roles.
Enable Authentication (page 257) Describes the process for enabling authentication for MongoDB deployments.
Create a User Administrator (page 258) Create users with special permissions to to create, modify, and remove other
users, as well as administer authentication credentials (e.g. passwords).
Add a User to a Database (page 259) Create non-administrator users using MongoDBs role-based authentication
system.
Change a Users Password (page 260) Only user administrators can edit credentials. This tutorial describes the process for editing an existing users password.
Generate a Key File (page 261) Use key file to allow the components of MongoDB sharded cluster or replica set to
mutually authenticate.
Deploy MongoDB with Kerberos Authentication (page 261) Describes the process, for MongoDB Enterprise, used
to enable and implement a Kerberos-based authentication system for MongoDB deployments.
Enable Authentication
Enable authentication using the auth or keyFile settings. Use auth for standalone instances, and keyFile
with replica sets and sharded clusters. keyFile implies auth and allows members of a MongoDB deployment to
authenticate internally.
Authentication requires at least one administrator user in the admin database. You can create the user before enabling
authentication or after enabling authentication.
See also:
Deploy MongoDB with Kerberos Authentication (page 261).
Procedures

You can enable authentication using either of the following procedures:

Create the Administrator Credentials and then Enable Authentication
1. Start the mongod or mongos instance without the auth or keyFile setting.
2. Create the administrator user as described in Create a User Administrator (page 258).
3. Re-start the mongod or mongos instance with the auth or keyFile setting.

5.3. Security Tutorials

257

MongoDB Documentation, Release 2.4.9

Enable Authentication and then Create Administrator

1. Start the mongod or mongos instance with the auth or keyFile setting.
2. Connect to the instance on the same system so that you can authenticate using the localhost exception (page 259).
3. Create the administrator user as described in Create a User Administrator (page 258).
Query Authenticated Users

If you have the userAdmin (page 269) or userAdminAnyDatabase (page 271) role on a database, you can query
authenticated users in that database with the following operation:
db.system.users.find()

Create a User Administrator

In a MongoDB deployment, users with either the userAdmin (page 269) or userAdminAnyDatabase (page 271)
roles are effective administrative superusers. Users with either of these roles can create and modify any other users
and can assign them any privileges. The user also can grant itself any privileges. In production deployments, this user
should have no other roles and should only administer users and privileges.
This should be the first user created for a MongoDB deployment. This user can then create all other users in the
system.
Important: The userAdminAnyDatabase (page 271) user can grant itself and any other user full access to the
entire MongoDB instance. The credentials to log in as this user should be carefully controlled.
Users with the userAdmin (page 269) and userAdminAnyDatabase (page 271) privileges are not the same as
the UNIX root superuser in that this role confers no additional access beyond user administration. These users
cannot perform administrative operations or read or write data without first conferring themselves with additional
permissions.
Note: The userAdmin (page 269) role is a database-specific privilege, and only grants a user the ability to administer
users on a single database. However, for the admin database, userAdmin (page 269) allows a user the ability to
gain userAdminAnyDatabase (page 271). Thus, for the admin database only, these roles are effectively the
same.

Create a User Administrator

1. Connect to the mongod or mongos by either:

Authenticating as an existing user with the userAdmin (page 269) or userAdminAnyDatabase
(page 271) role.
Authenticating using the localhost exception (page 259). When creating the first user in a deployment, you
must authenticate using the localhost exception (page 259).
2. Switch to the admin database:
use admin

3. Add the user with either the userAdmin (page 269) role or userAdminAnyDatabase (page 271) role,
and only that role, by issuing a command similar to the following, where <username> is the username and
<password> is the password:

258

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

db.addUser( { user: "<username>",

pwd: "<password>",
roles: [ "userAdminAnyDatabase" ] } )

To authenticate as this user, you must authenticate against the admin database.
Authenticate with Full Administrative Access via Localhost

If there are no users for the admin database, you can connect with full administrative access via the localhost interface.
This bypass exists to support bootstrapping new deployments. This approach is useful, for example, if you want to run
mongod or mongos with authentication before creating your first user.
To authenticate via localhost, connect to the mongod or mongos from a client running on the same system. Your
connection will have full administrative access.
To disable the localhost bypass, set the enableLocalhostAuthBypass parameter using setParameter during startup:
mongod --setParameter enableLocalhostAuthBypass=0

Note: For versions of MongoDB 2.2 prior to 2.2.4, if mongos is running with keyFile, then all users connecting
over the localhost interface must authenticate, even if there arent any users in the admin database. Connections on
localhost are not correctly granted full access on sharded systems that run those versions.
MongoDB 2.2.4 resolves this issue.
Note: In version 2.2, you cannot add the first user to a sharded cluster using the localhost connection. If you are
running a 2.2 sharded cluster and want to enable authentication, you must deploy the cluster and add the first user to
the admin database before restarting the cluster to run with keyFile.

Add a User to a Database

To add a user to a database you must authenticate to that database as a user with the userAdmin (page 269) or
userAdminAnyDatabase (page 271) role. If you have not first created a user with one of those roles, do so as
described in Create a User Administrator (page 258).
When adding a user to multiple databases, you must define the user for each database. See Password Hashing Insecurity (page 275) for important security information.
To add a user, pass the db.addUser() method a well formed privilege document (page 267) that contains the users
credentials and privileges. The db.addUser() method adds the document to the databases system.users
(page 272) collection.
Changed in version 2.4: In previous versions of MongoDB, you could change an existing users password by calling
db.addUser() again with the users username and their updated password. Anything specified in the addUser()
method would override the existing information for that user. In newer versions of MongoDB, this will result in a
duplicate key error.
To change a users password in version 2.4 or newer, see Change a Users Password (page 260).
For the structure of a privilege document, see system.users (page 272). For descriptions of user roles, see User
Privilege Roles in MongoDB (page 267).
Example

5.3. Security Tutorials

259

MongoDB Documentation, Release 2.4.9

The following creates a user named Alice in the products database and gives her readWrite and dbAdmin
privileges.
use products
db.addUser( { user: "Alice",
pwd: "Moon1234",
roles: [ "readWrite", "dbAdmin" ]
} )

Example
The following creates a user named Bob in the admin database. The privilege document (page 272) uses
Bobs credentials from the products database and assigns him userAdmin privileges.
use admin
db.addUser( { user: "Bob",
userSource: "products",
roles: [ "userAdmin" ]
} )

Example
The following creates a user named Carlos in the admin database and gives him readWrite access to the
config database, which lets him change certain settings for sharded clusters, such as to disable the balancer.
db = db.getSiblingDB('admin')
db.addUser( { user: "Carlos",
pwd: "Moon1234",
roles: [ "clusterAdmin" ],
otherDBRoles: { config: [ "readWrite" ]
} } )

Only the admin database supports the otherDBRoles (page 274) field.

Change a Users Password

New in version 2.4.
To change a users password, you must have the userAdmin (page 269) role on the database that contains the
definition of the user whose password you wish to change.
To update the password, pass the
db.changeUserPassword() method.

users

username

and

the

new

desired

password

to

the

Example
The following operation changes the reporting users password to SOhSS3TbYhxusooLiW8ypJPxmt1oOfL:
db = db.getSiblingDB('records')
db.changeUserPassword("reporting", "SOhSS3TbYhxusooLiW8ypJPxmt1oOfL")

Note: In previous versions of MongoDB, you could change an existing users password by calling db.addUser()
again with the users username and their updated password. Anything specified in the addUser() method would
override the existing information for that user. In newer versions of MongoDB, this will result in a duplicate key error.
For more about changing a users password prior to version 2.4, see: Add a User to a Database (page 259).
260

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

Generate a Key File

This section describes how to generate a key file to store authentication information. After generating a key file,
specify the key file using the keyFile option when starting a mongod or mongos instance.
A keys length must be between 6 and 1024 characters and may only contain characters in the base64 set. The key
file must not have group or world permissions on UNIX systems. Key file permissions are not checked on Windows
systems.
Generate a Key File

Use the following openssl command at the system shell to generate pseudo-random content for a key file:
openssl rand -base64 741

Note: Key file permissions are not checked on Windows systems.

Key File Properties

Be aware that MongoDB strips whitespace characters (e.g. x0d, x09, and x20) for cross-platform convenience. As
a result, the following operations produce identical keys:
echo
echo
echo
echo

-e
-e
-e
-e

"my secret key" > key1

"my secret key\n" > key2
"my
secret
key" > key3
"my\r\nsecret\r\nkey\r\n" > key4

Deploy MongoDB with Kerberos Authentication

New in version 2.4.
MongoDB Enterprise supports authentication using a Kerberos service. Kerberos is an industry standard authentication
protocol for large client/server system. With Kerberos MongoDB and application ecosystems can take advantage of
existing authentication infrastructure and processes.
Setting up and configuring a Kerberos deployment is beyond the scope of this document. In order to use MongoDB
with Kerberos, you must have a properly configured Kerberos deployment and the ability to generate a valid keytab
file for each mongod instance in your MongoDB deployment.
Note:
The following assumes that you have a valid Kerberos keytab file for your realm accessible on your system.
The examples below assume that the keytab file is valid and is located at
http://docs.mongodb.org/manualopt/mongodb/mongod.keytab and is only accessible to the user
that runs the mongod process.

Process Overview

To run MongoDB with Kerberos support, you must:

Configure a Kerberos service principal for each mongod and mongos instance in your MongoDB deployment.

5.3. Security Tutorials

261

MongoDB Documentation, Release 2.4.9

Generate and distribute keytab files for each MongoDB component (i.e. mongod and mongos)in your deployment. Ensure that you only transmit keytab files over secure channels.
Optional. Start the mongod instance without auth and create users inside of MongoDB that you can use to
bootstrap your deployment.
Start mongod and mongos with the KRB5_KTNAME environment variable as well as a number of required run
time options.
If you did not create Kerberos user accounts, you can use the localhost exception (page 259) to create users at
this point until you create the first user on the admin database.
Authenticate clients, including the mongo shell using Kerberos.
Operations

Create Users and Privilege Documents For every user that you want to be able to authenticate using Kerberos, you
must create corresponding privilege documents in the system.users (page 272) collection to provision access to
users. Consider the following document:
{
user: "application/[email protected]",
roles: ["read"],
userSource: "$external"
}

This grants the Kerberos user principal application/[email protected] read only access to a
database. The userSource (page 273) $external reference allows mongod to consult an external source (i.e.
Kerberos) to authenticate this user.
In the mongo shell you can pass the db.addUser() a user privilege document to provision access to users, as in
the following operation:
db = db.getSiblingDB("records")
db.addUser( {
"user": "application/[email protected]",
"roles": [ "read" ],
"userSource": "$external"
} )

These operations grants the Kerberos user application/[email protected] access to the records
database.
To remove access to a user, use the remove() method, as in the following example:
db.system.users.remove( { user: "application/[email protected]" } )

To modify a user document, use update (page 42) operations on documents in the system.users (page 272)
collection.
See also:
system.users Privilege Documents (page 272) and User Privilege Roles in MongoDB (page 267).
Start mongod with Kerberos Support Once you have provisioned privileges to users in the mongod, and obtained
a valid keytab file, you must start mongod using a command in the following form:
env KRB5_KTNAME=<path to keytab file> <mongod invocation>

262

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

For successful operation with mongod use the following run time options in addition to your normal default configuration options:
--setParameter with the authenticationMechanisms=GSSAPI argument to enable support for
Kerberos.
--auth to enable authentication.
--keyFile to allow components of a single MongoDB deployment to communicate with each other, if needed
to support replica set and sharded cluster operations. keyFile implies auth.
For example, consider the following invocation:
env KRB5_KTNAME=/opt/mongodb/mongod.keytab \
/opt/mongodb/bin/mongod --dbpath /opt/mongodb/data \
--fork --logpath /opt/mongodb/log/mongod.log \
--auth --setParameter authenticationMechanisms=GSSAPI

You can also specify these options using the configuration file. As in the following:
# /opt/mongodb/mongod.conf, Example configuration file.
fork = true
auth = true
dbpath = /opt/mongodb/data
logpath = /opt/mongodb/log/mongod.log
setParameter = authenticationMechanisms=GSSAPI

To use this configuration file, start mongod as in the following:

env KRB5_KTNAME=/opt/mongodb/mongod.keytab \
/opt/mongodb/bin/mongod --config /opt/mongodb/mongod.conf

To start a mongos instance using Kerberos, you must create a Kerberos service principal and deploy a keytab file for
this instance, and then start the mongos with the following invocation:
env KRB5_KTNAME=/opt/mongodb/mongos.keytab \
/opt/mongodb/bin/mongos
--configdb shard0.example.net,shard1.example.net,shard2.example.net \
--setParameter authenticationMechanisms=GSSAPI \
--keyFile /opt/mongodb/mongos.keyfile

Tip
If you installed MongoDB Enterprise using one of the official .deb or .rpm packages and are controlling the
mongod instance using the included init/upstart scripts, you can set the KR5_KTNAME variable in the default environment settings file. For .rpm packages this file is located at /etc/sysconfig/mongod. For .deb packages,
this file is /etc/default/mongodb. Set the value in a line that resembles the following:
export KRB5_KTNAME="<setting>"

If you encounter problems when trying to start mongod or mongos, please see the troubleshooting section (page 264)
for more information.
Important: Before users can authenticate to MongoDB using Kerberos you must create users (page 262) and grant
them privileges within MongoDB. If you have not created users when you start MongoDB with Kerberos you can
use the localhost authentication exception (page 259) to add users. See the Create Users and Privilege Documents
(page 262) section and the User Privilege Roles in MongoDB (page 267) document for more information.

5.3. Security Tutorials

263

MongoDB Documentation, Release 2.4.9

Authenticate mongo Shell with Kerberos To connect to a mongod instance using the mongo shell you must begin
by using the kinit program to initialize and authenticate a Kerberos session. Then, start a mongo instance, and use
the db.auth() method, to authenticate against the special $external database, as in the following operation:
use $external
db.auth( { mechanism: "GSSAPI", user: "application/[email protected]" } )

Alternately, you can authenticate using command line options to mongo, as in the following equivalent example:
mongo --authenticationMechanism=GSSAPI
--authenticationDatabase='$external' \
--username application/[email protected]

These operations authenticate the Kerberos principal name application/[email protected] to the

connected mongod, and will automatically acquire all available privileges as needed.
Use MongoDB Drivers to Authenticate with Kerberos At the time of release, the C++, Java, C#, and Python
drivers all provide support for Kerberos authentication to MongoDB. Consider the following tutorials for more information:
Authenticate to MongoDB with the Java Driver22
Authenticate to MongoDB with the C# Driver23
Authenticate to MongoDB with the C++ Driver24
Python Authentication Examples25
Kerberos and the HTTP Console MongoDB does not support kerberizing the HTTP Console26 .
Troubleshooting

Kerberos Configuration Checklist If youre having trouble getting mongod to start with Kerberos, there are a
number of Kerberos-specific issues that can prevent successful authentication. As you begin troubleshooting your
Kerberos deployment, ensure that:
The mongod is from MongoDB Enterprise.
You are not using the HTTP Console27 . MongoDB Enterprise does not support Kerberos authentication over the
HTTP Console interface.
You have a valid keytab file specified in the environment running the mongod. For the mongod instance
running on the db0.example.net host, the service principal should be mongodb/db0.example.net.
DNS allows the mongod to resolve the components of the Kerberos infrastructure. You should have both A and
PTR records (i.e. forward and reverse DNS) for the system that runs the mongod instance.
The canonical system hostname of the system that runs the mongod instance is the resolvable fully qualified
domain for this host. Test system hostname resolution with the hostname -f command at the system prompt.
Both the Kerberos KDC and the system running mongod instance must be able to resolve each other using DNS
28
22 http://docs.mongodb.org/ecosystem/tutorial/authenticate-with-java-driver/
23 http://docs.mongodb.org/ecosystem/tutorial/authenticate-with-csharp-driver/
24 http://docs.mongodb.org/ecosystem/tutorial/authenticate-with-cpp-driver/
25 http://api.mongodb.org/python/current/examples/authentication.html
26 http://docs.mongodb.org/ecosystem/tools/http-interface/#http-console
27 http://docs.mongodb.org/ecosystem/tools/http-interface/#http-console
28

By default, Kerberos attempts to resolve hosts using the content of the /etc/kerb5.conf before using DNS to resolve hosts.

264

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

The time systems of the systems running the mongod instances and the Kerberos infrastructure are synchronized. Time differences greater than 5 minutes will prevent successful authentication.
If you still encounter problems with Kerberos, you can start both mongod and mongo (or another client) with the
environment variable KRB5_TRACE set to different files to produce more verbose logging of the Kerberos process to
help further troubleshooting, as in the following example:
env KRB5_KTNAME=/opt/mongodb/mongod.keytab \
KRB5_TRACE=/opt/mongodb/log/mongodb-kerberos.log \
/opt/mongodb/bin/mongod --dbpath /opt/mongodb/data \
--fork --logpath /opt/mongodb/log/mongod.log \
--auth --setParameter authenticationMechanisms=GSSAPI

Common Error Messages In some situations, MongoDB will return error messages from the GSSAPI interface if
there is a problem with the Kerberos service.
GSSAPI error in client while negotiating security context.
This error occurs on the client and reflects insufficient credentials or a malicious attempt to authenticate.
If you receive this error ensure that youre using the correct credentials and the correct fully qualified
domain name when connecting to the host.
GSSAPI error acquiring credentials.
This error only occurs when attempting to start the mongod or mongos and reflects improper configuration of system hostname or a missing or incorrectly configured keytab file. If you encounter this problem,
consider all the items in the Kerberos Configuration Checklist (page 264), in particular:
examine the keytab file, with the following command:
klist -k <keytab>

Replace <keytab> with the path to your keytab file.

check the configured hostname for your system, with the following command:
hostname -f

Ensure that this name matches the name in the keytab file, or use the saslHostName to pass
MongoDB the correct hostname.
Enable the Traditional MongoDB Authentication Mechanism For testing and development purposes you can
enable both the Kerberos (i.e. GSSAPI) authentication mechanism in combination with the traditional MongoDB
challenge/response authentication mechanism (i.e. MONGODB-CR), using the following setParameter run-time
option:
mongod --setParameter authenticationMechanisms=GSSAPI,MONGODB-CR

Warning: All keyFile internal authentication between members of a replica set or sharded cluster still uses
the MONGODB-CR authentication mechanism, even if MONGODB-CR is not enabled. All client authentication will
still use Kerberos.

5.3.3 Create a Vulnerability Report

If you believe you have discovered a vulnerability in MongoDB or have experienced a security incident related to
MongoDB, please report the issue so it can be avoided in the future.

5.3. Security Tutorials

265

MongoDB Documentation, Release 2.4.9

To report an issue, we strongly suggest filing a ticket in our Security project in JIRA
<https://jira.mongodb.org/browse/SECURITY/>_ . MongoDB, Inc responds to vulnerability notifications within 48
hours.
Create the Report in JIRA
Submit a ticket in the Security29 project at: https://jira.mongodb.org/browse/SECURITY/. The ticket number will
become the reference identification for the issue for its lifetime. You can use this identifier for tracking purposes.
Information to Provide
All vulnerability reports should contain as much information as possible so MongoDBs developers can move quickly
to resolve the issue. In particular, please include the following:
The name of the product.
Common Vulnerability information, if applicable, including:
CVSS (Common Vulnerability Scoring System) Score.
CVE (Common Vulnerability and Exposures) Identifier.
Contact information, including an email address and/or phone number, if applicable.
Send the Report via Email
While JIRA is the preferred reporting method, you may also report vulnerabilities via email to [email protected] .
You may encrypt email using MongoDBs public key at http://docs.mongodb.org/10gen-security-gpg-key.asc.
MongoDB, Inc. responds to vulnerability reports sent via email with a response email that contains a reference number
for a JIRA ticket posted to the SECURITY31 project.
Evaluation of a Vulnerability Report
MongoDB, Inc. validates all submitted vulnerabilities and uses Jira to track all communications regarding a vulnerability, including requests for clarification or additional information. If needed, MongoDB representatives set up a
conference call to exchange information regarding the vulnerability.
Disclosure
MongoDB, Inc. requests that you do not publicly disclose any information regarding the vulnerability or exploit the
issue until it has had the opportunity to analyze the vulnerability, to respond to the notification, and to notify key users,
customers, and partners.
The amount of time required to validate a reported vulnerability depends on the complexity and severity of the issue.
MongoDB, Inc. takes all required vulnerabilities very seriously and will always ensure that there is a clear and open
channel of communication with the reporter.
29 https://jira.mongodb.org/browse/SECURITY
30 [email protected]
31 https://jira.mongodb.org/browse/SECURITY

266

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

After validating an issue, MongoDB, Inc. coordinates public disclosure of the issue with the reporter in a mutually
agreed timeframe and format. If required or requested, the reporter of a vulnerability will receive credit in the published
security bulletin.

5.4 Security Reference

5.4.1 Security Methods in the mongo Shell
Name
db.addUser()

Description
Adds a user to a database, and allows administrators to configure the users
privileges.
db.auth()
Authenticates a user to a database.
db.changeUserPassword() Changes an existing users password.

5.4.2 Security Reference Documentation

User Privilege Roles in MongoDB (page 267) Reference on user privilege roles and corresponding access.
system.users Privilege Documents (page 272) Reference on documents used to store user credentials and privilege
roles.
Default MongoDB Port (page 275) List of default ports used by MongoDB.
User Privilege Roles in MongoDB
New in version 2.4: In version 2.4, MongoDB adds support for the following user roles:
Roles

Changed in version 2.4.

Roles in MongoDB provide users with a set of specific privileges, on specific logical databases. Users may have
multiple roles and may have different roles on different logical database. Roles only grant privileges and never limit
access: if a user has read (page 268) and readWriteAnyDatabase (page 271) permissions on the records
database, that user will be able to write data to the records database.
Note:
By default, MongoDB 2.4 is backwards-compatible with the MongoDB 2.2 access control roles.
You can explicitly disable this backwards-compatibility by setting the
supportCompatibilityFormPrivilegeDocuments option to 0 during startup, as in the following
command-line invocation of MongoDB:
mongod --setParameter supportCompatibilityFormPrivilegeDocuments=0

In general, you should set this option if your deployment does not need to support legacy user documents. Typically
legacy user documents are only useful during the upgrade process and while you migrate applications to the updated
privilege document form.
See privilege documents (page 272) and Delegated Credentials for MongoDB Authentication (page 274) for more
information about permissions and authentication in MongoDB.

5.4. Security Reference

267

MongoDB Documentation, Release 2.4.9

Database User Roles

read
Provides users with the ability to read data from any collection within a specific logical database. This includes
find() and the following database commands:
aggregate
checkShardingIndex
cloneCollectionAsCapped
collStats
count
dataSize
dbHash
dbStats
distinct
filemd5
geoNear
geoSearch
geoWalk
group
mapReduce (inline output only.)
text (beta feature.)
readWrite
Provides users with the ability to read from or write to any collection within a specific logical database. Users
with readWrite (page 268) have access to all of the operations available to read (page 268) users, as well
as the following basic write operations: insert(), remove(), and update().
Additionally, users with the readWrite (page 268) have access to the following database commands:
cloneCollection (as the target database.)
convertToCapped
create (and to create collections implicitly.)
drop()
dropIndexes
emptycapped
ensureIndex()
findAndModify
mapReduce (output to a collection.)
renameCollection (within the same database.)

268

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

Database Administration Roles

dbAdmin
Provides the ability to perform the following set of administrative operations within the scope of this logical
database.
clean
collMod
collStats
compact
convertToCapped
create
db.createCollection()
dbStats
drop()
dropIndexes
ensureIndex()
indexStats
profile
reIndex
renameCollection (within a single database.)
validate
Furthermore, only dbAdmin (page 269) has the ability to read the system.profile (page 227) collection.
userAdmin
Allows users to read and write data to the system.users (page 272) collection of the users database.
Users with this role will be able to modify permissions for existing users and create new users. userAdmin
(page 269) does not restrict the permissions that a user can grant, and a userAdmin (page 269) user can grant
privileges to themselves or other users in excess of the userAdmin (page 269) users current privileges.
Important: userAdmin (page 269) is effectively the superuser role for a specific database. Users with
userAdmin (page 269) can grant themselves all privileges. However, userAdmin (page 269) does not explicitly authorize a user for any privileges beyond user administration.
Note: The userAdmin (page 269) role is a database-specific privilege, and only grants a user the ability to
administer users on a single database. However, for the admin database, userAdmin (page 269) allows a user
the ability to gain userAdminAnyDatabase (page 271). Thus, for the admin database only, these roles are
effectively the same.

Administrative Roles

clusterAdmin
clusterAdmin (page 269) grants access to several administration operations that affect or present information
about the whole system, rather than just a single database. These privileges include but are not limited to replica
set and sharded cluster administrative functions.

5.4. Security Reference

269

MongoDB Documentation, Release 2.4.9

clusterAdmin (page 269) is only applicable on the admin database, and does not confer any access to the
local or config databases.
Specifically, users with the clusterAdmin (page 269) role have access to the following operations:
addShard
closeAllDatabases
connPoolStats
connPoolSync
_cpuProfilerStart
_cpuProfilerStop
cursorInfo
diagLogging
dropDatabase
enableSharding
flushRouterConfig
fsync
db.fsyncUnlock()
getCmdLineOpts
getLog
getParameter
getShardMap
getShardVersion
hostInfo
db.currentOp()
db.killOp()
listDatabases
listShards
logRotate
moveChunk
movePrimary
netstat
removeShard
repairDatabase
replSetFreeze
replSetGetStatus
replSetInitiate
replSetMaintenance

270

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

replSetReconfig
replSetStepDown
replSetSyncFrom
resync
serverStatus
setParameter
setShardVersion
shardCollection
shardingState
shutdown
splitChunk
splitVector
split
top
touch
unsetSharding
For some cluster administration operations, MongoDB requires read and write access to the local or config
databases. You must specify this access separately from clusterAdmin (page 269). See the Combined Access
(page 272) section for more information.
Any Database Roles

Note: You must specify the following any database roles on the admin databases. These roles apply to all
databases in a mongod instance and are roughly equivalent to their single-database equivalents.
If you add any of these roles to a user privilege document (page 272) outside of the admin database, the privilege
will have no effect. However, only the specification of the roles must occur in the admin database, with delegated
authentication credentials (page 274), users can gain these privileges by authenticating to another database.
readAnyDatabase
readAnyDatabase (page 271) provides users with the same read-only permissions as read (page 268),
except it applies to all logical databases in the MongoDB environment.
readWriteAnyDatabase
readWriteAnyDatabase (page 271) provides users with the same read and write permissions as
readWrite (page 268), except it applies to all logical databases in the MongoDB environment.
userAdminAnyDatabase
userAdminAnyDatabase (page 271) provides users with the same access to user administration operations
as userAdmin (page 269), except it applies to all logical databases in the MongoDB environment.
Important: Because users with userAdminAnyDatabase (page 271) and userAdmin (page 269) have
the ability to create and modify permissions in addition to their own level of access, this role is effectively the
MongoDB system superuser. However, userAdminAnyDatabase (page 271) and userAdmin (page 269)
do not explicitly authorize a user for any privileges beyond user administration.

5.4. Security Reference

271

MongoDB Documentation, Release 2.4.9

dbAdminAnyDatabase
dbAdminAnyDatabase (page 271) provides users with the same access to database administration operations
as dbAdmin (page 269), except it applies to all logical databases in the MongoDB environment.
Combined Access

Some operations are only available to users that have multiple roles. Consider the following:
sh.status() Requires clusterAdmin (page 269) and read (page 268) access to the config (page 548)
database.
applyOps, eval 32 Requires readWriteAnyDatabase (page 271), userAdminAnyDatabase (page 271),
dbAdminAnyDatabase (page 271) and clusterAdmin (page 269) (on the admin database.)
Some operations related to cluster administration are not available to users who only have the clusterAdmin
(page 269) role:
rs.conf() Requires read (page 268) on the local database.
sh.addShard() Requires readWrite (page 268) on the config database.
system.users Privilege Documents
Changed in version 2.4.
Overview

The documents in the <database>.system.users (page 272) collection store credentials and user privilege
information used by the authentication system to provision access to users in the MongoDB system. See User Privilege
Roles in MongoDB (page 267) for more information about access roles, and Security (page 237) for an overview of
security in MongoDB.
Data Model

<database>.system.users
Changed in version 2.4.
Documents in the <database>.system.users (page 272) collection stores credentials and user roles
(page 267) for users who have access to the database. Consider the following prototypes of user privilege
documents:
{
user: "<username>",
pwd: "<hash>",
roles: []
}
{
user: "<username>",
userSource: "<database>",
roles: []
}

272

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

Note: The pwd (page 273) and userSource (page 273) fields are mutually exclusive. A single document
cannot contain both.
The following privilege document with the otherDBRoles (page 274) field is only supported on the admin
database:
{
user: "<username>",
userSource: "<database>",
otherDBRoles: {
<database0> : [],
<database1> : []
},
roles: []
}

Consider the content of the following fields in the system.users (page 272) documents:
<database>.system.users.user
user (page 273) is a string that identifies each user. Users exist in the context of a single logical database;
however, users from one database may obtain access in another database by way of the otherDBRoles
(page 274) field on the admin database, the userSource (page 273) field, or the Any Database Roles
(page 271).
<database>.system.users.pwd
pwd (page 273) holds a hashed shared secret used to authenticate the user (page 273). pwd (page 273)
field is mutually exclusive with the userSource (page 273) field.
<database>.system.users.roles
roles (page 273) holds an array of user roles. The available roles are:
read (page 268)
readWrite (page 268)
dbAdmin (page 269)
userAdmin (page 269)
clusterAdmin (page 269)
readAnyDatabase (page 271)
readWriteAnyDatabase (page 271)
userAdminAnyDatabase (page 271)
dbAdminAnyDatabase (page 271)
See Roles (page 267) for full documentation of all available user roles.
<database>.system.users.userSource
A string that holds the name of the database that contains the credentials for the user. If userSource
(page 273) is $external, then MongoDB will use an external resource, such as Kerberos, for authentication credentials.
Note: In the current release, the only external authentication source is Kerberos, which is only available
in MongoDB Enterprise.
Use userSource (page 273) to ensure that a single users authentication credentials are only stored in a
single location in a mongod instances data.

5.4. Security Reference

273

MongoDB Documentation, Release 2.4.9

A userSource (page 273) and user (page 273) pair identifies a unique user in a MongoDB system.
admin.system.users.otherDBRoles
A document that holds one or more fields with a name that is the name of a database in the MongoDB
instance with a value that holds a list of roles this user has on other databases. Consider the following
example:
{
user: "admin",
userSource: "$external",
roles: [ "clusterAdmin"],
otherDBRoles:
{
config: [ "read" ],
records: [ "dbAdmin" ]
}
}

This user has the following privileges:

clusterAdmin (page 269) on the admin database,
read (page 268) on the config (page 548) database, and
dbAdmin (page 269) on the records database.
Delegated Credentials for MongoDB Authentication

New in version 2.4.

With a new document format in the system.users (page 272) collection, MongoDB now supports the ability
to delegate authentication credentials to other sources and databases. The userSource (page 273) field in these
documents forces MongoDB to use another source for credentials.
Consider the following document in a system.users (page 272) collection in a database named accounts:
{
user: "application0",
pwd: "YvuolxMtaycghk2GMrzmImkG4073jzAw2AliMRul",
roles: []
}

Then for every database that the application0 user requires access, add documents to the system.users
(page 272) collection that resemble the following:
{
user: "application0",
roles: ['readWrite'],
userSource: "accounts"
}

To gain privileges to databases where the application0 has access, you must first authenticate to the accounts
database.
Disable Legacy Privilege Documents

By default MongoDB 2.4 includes support for both new, role-based privilege documents style as well 2.2 and earlier
privilege documents. MongoDB assumes any privilege document without a roles (page 273) field is a 2.2 or earlier
document.
274

Chapter 5. Security

MongoDB Documentation, Release 2.4.9

To ensure that mongod instances will only provide access to users defined with the new role-based privilege documents, use the following setParameter run-time option:
mongod --setParameter supportCompatibilityFormPrivilegeDocuments=0

Default MongoDB Port

The following table lists the default ports used by MongoDB:
Default
Port
27017
27018
27019
28017

Description
The default port for mongod and mongos instances. You can change this port with port or
--port.
The default port when running with --shardsvr runtime operation or shardsvr setting.
The default port when running with --configsvr runtime operation or configsvr setting.
The default port for the web status page. The web status page is always accessible at a port number
that is 1000 greater than the port determined by port.

5.4.3 Security Release Notes Alerts

Security Release Notes (page 275) Security vulnerability for password.
Security Release Notes
Access to system.users Collection

Changed in version 2.4.

In 2.4, only users with the userAdmin role have access to the system.users collection.
In version 2.2 and earlier, the read-write users of a database all have access to the system.users collection, which
contains the user names and user password hashes. 33
Password Hashing Insecurity

If a user has the same password for multiple databases, the hash will be the same. A malicious user could exploit this
to gain access on a second database using a different users credentials.
As a result, always use unique username and password combinations for each database.
Thanks to Will Urbanski, from Dell SecureWorks, for identifying this issue.

33

Read-only users do not have access to the system.users collection.

5.4. Security Reference

275

MongoDB Documentation, Release 2.4.9

276

Chapter 5. Security

CHAPTER 6

Aggregation

Aggregations operations process data records and return computed results. Aggregation operations group values from
multiple documents together, and can perform a variety of operations on the grouped data to return a single result.
MongoDB provides three ways to perform aggregation: the aggregation pipeline (page 281), the map-reduce function
(page 284), and single purpose aggregation methods and commands (page 286).
Aggregation Introduction (page 277) A high-level introduction to aggregation.
Aggregation Concepts (page 281) Introduces the use and operation of the data aggregation modalities available in
MongoDB.
Aggregation Pipeline (page 281) The aggregation pipeline is a framework for performing aggregation tasks,
modeled on the concept of data processing pipelines. Using this framework, MongoDB passes the documents of a single collection through a pipeline. The pipeline transforms the documents into aggregated
results, and is accessed through the aggregate database command.
Map-Reduce (page 284) Map-reduce is a generic multi-phase data aggregation modality for processing quantities of data. MongoDB provides map-reduce with the mapReduce database command.
Single Purpose Aggregation Operations (page 286) MongoDB provides a collection of specific data aggregation operations to support a number of common data aggregation functions. These operations include
returning counts of documents, distinct values of a field, and simple grouping operations.
Aggregation Mechanics (page 289) Details internal optimization operations, limits, support for sharded collections, and concurrency concerns.
Aggregation Examples (page 292) Examples and tutorials for data aggregation operations in MongoDB.
Aggregation Reference (page 308) References for all aggregation operations material for all data aggregation methods in MongoDB.

6.1 Aggregation Introduction

Aggregations are operations that process data records and return computed results. MongoDB provides a rich set
of aggregation operations that examine and perform calculations on the data sets. Running data aggregation on the
mongod instance simplifies application code and limits resource requirements.
Like queries, aggregation operations in MongoDB use collections of documents as an input and return results in the
form of one or more documents.

277

MongoDB Documentation, Release 2.4.9

6.1.1 Aggregation Modalities

Aggregation Pipelines
MongoDB 2.2 introduced a new aggregation framework (page 281), modeled on the concept of data processing
pipelines. Documents enter a multi-stage pipeline that transforms the documents into an aggregated result.
The most basic pipeline stages provide filters that operate like queries and document transformations that modify the
form of the output document.
Other pipeline operations provide tools for grouping and sorting documents by specific field or fields as well as tools
for aggregating the contents of arrays, including arrays of documents. In addition, pipeline stages can use operators
for tasks such as calculating the average or concatenating a string.
The pipeline provides efficient data aggregation using native operations within MongoDB, and is the preferred method
for data aggregation in MongoDB.

Figure 6.1: Diagram of the annotated aggregation pipeline operation. The aggregation pipeline has two phases:
$match and $group.

Map-Reduce
MongoDB also provides map-reduce (page 284) operations to perform aggregation. In general, map-reduce operations
have two phases: a map stage that processes each document and emits one or more objects for each input document,
278

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

and reduce phase that combines the output of the map operation. Optionally, map-reduce can have a finalize stage to
make final modifications to the result. Like other aggregation operations, map-reduce can specify a query condition to
select the input documents as well as sort and limit the results.
Map-reduce uses custom JavaScript functions to perform the map and reduce operations, as well as the optional finalize
operation. While the custom JavaScript provide great flexibility compared to the aggregation pipeline, in general, mapreduce is less efficient and more complex than the aggregation pipeline.
Additionally, map-reduce operations can have output sets that exceed the 16 megabyte output limitation of the aggregation pipeline.
Note: Starting in MongoDB 2.4, certain mongo shell functions and properties are inaccessible in map-reduce operations. MongoDB 2.4 also provides support for multiple JavaScript operations to run at the same time. Before
MongoDB 2.4, JavaScript code executed in a single thread, raising concurrency issues for map-reduce.

Figure 6.2: Diagram of the annotated map-reduce operation.

Single Purpose Aggregation Operations

For a number of common single purpose aggregation operations (page 286), MongoDB provides special purpose
database commands. These common aggregation operations are: returning a count of matching documents, returning
the distinct values for a field, and grouping data based on the values of a field. All of these operations aggregate
documents from a single collection. While these operations provide simple access to common aggregation processes,
they lack the flexibility and capabilities of the aggregation pipeline and map-reduce.

6.1. Aggregation Introduction

279

MongoDB Documentation, Release 2.4.9

Figure 6.3: Diagram of the annotated distinct operation.

280

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

6.1.2 Additional Features and Behaviors

Both the aggregation pipeline and map-reduce can operate on a sharded collection (page 479). Map-reduce operations
can also output to a sharded collection. See Aggregation Pipeline and Sharded Collections (page 291) and Map-Reduce
and Sharded Collections (page 291) for details.
The aggregation pipeline can use indexes to improve its performance during some of its stages. In addition, the aggregation pipeline has an internal optimization phase. See Pipeline Operators and Indexes (page 283) and Aggregation
Pipeline Optimization (page 289) for details.
For a feature comparison of the aggregation pipeline, map-reduce, and the special group functionality, see Aggregation
Commands Comparison (page 308).

6.2 Aggregation Concepts

MongoDB provides the three approaches to aggregation, each with its own strengths and purposes for a given situation.
This section describes these approaches and also describes behaviors and limitations specific to each approach. See
also the chart (page 308) that compares the approaches.
Aggregation Pipeline (page 281) The aggregation pipeline is a framework for performing aggregation tasks, modeled
on the concept of data processing pipelines. Using this framework, MongoDB passes the documents of a single
collection through a pipeline. The pipeline transforms the documents into aggregated results, and is accessed
through the aggregate database command.
Map-Reduce (page 284) Map-reduce is a generic multi-phase data aggregation modality for processing quantities of
data. MongoDB provides map-reduce with the mapReduce database command.
Single Purpose Aggregation Operations (page 286) MongoDB provides a collection of specific data aggregation operations to support a number of common data aggregation functions. These operations include returning counts
of documents, distinct values of a field, and simple grouping operations.
Aggregation Mechanics (page 289) Details internal optimization operations, limits, support for sharded collections,
and concurrency concerns.

6.2.1 Aggregation Pipeline

New in version 2.2.
The aggregation pipeline is a framework for data aggregation modeled on the concept of data processing pipelines.
Documents enter a multi-stage pipeline that transforms the documents into an aggregated results.
The aggregation pipeline provides an alternative to map-reduce and may be the preferred solution for many aggregation
tasks where the complexity of map-reduce may be unwarranted.
Aggregation pipeline have some limitations on value types and result size. See Aggregation Pipeline Limits (page 290)
for details on limits and restrictions on the aggregation pipeline.
Pipeline
Conceptually, documents from a collection travel through an aggregation pipeline, which transforms these objects as
they pass through. For those familiar with UNIX-like shells (e.g. bash), the concept is analogous to the pipe (i.e. |).
The MongoDB aggregation pipeline starts with the documents of a collection and streams the documents from one
pipeline operator to the next to process the documents. Each operator in the pipeline transforms the documents as they
pass through the pipeline. Pipeline operators do not need to produce one output document for every input document.
Operators may generate new documents or filter out documents. Pipeline operators can be repeated in the pipeline.
6.2. Aggregation Concepts

281

MongoDB Documentation, Release 2.4.9

Figure 6.4: Diagram of the annotated aggregation pipeline operation. The aggregation pipeline has two phases:
$match and $group.

282

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

Important: The result of aggregation pipeline is a document and is subject to the BSON Document size limit, which
is currently 16 megabytes.
See aggregation-pipeline-operator-reference for the list of pipeline operators that define the stages.
For example usage of the aggregation pipeline, consider Aggregation with User Preference Data (page 296)
and Aggregation with the Zip Code Data Set (page 293), as well as the aggregate command and the
db.collection.aggregate() method reference pages.
Pipeline Expressions
Each pipeline operator takes a pipeline expression as its operand. Pipeline expressions specify the transformation to
apply to the input documents. Expressions have a document structure and can contain fields, values, and operators.
Pipeline expressions can only operate on the current document in the pipeline and cannot refer to data from other
documents: expression operations provide in-memory transformation of documents.
Generally, expressions are stateless and are only evaluated when seen by the aggregation process with one exception:
accumulator expressions. The accumulator expressions, used with the $group pipeline operator, maintain their state
(e.g. totals, maximums, minimums, and related data) as documents progress through the pipeline.
For the expression operators, see aggregation-expression-operators.
Aggregation Pipeline Behavior
In MongoDB, the aggregate command operates on a single collection, logically passing the entire collection into
the aggregation pipeline. To optimize the operation, wherever possible, use the following strategies to avoid scanning
the entire collection.
Pipeline Operators and Indexes

The $match, $sort, $limit, and $skip pipeline operators can take advantage of an index when they occur at the
beginning of the pipeline before any of the following aggregation operators: $project, $unwind, and $group.
New in version 2.4: The $geoNear pipeline operator takes advantage of a geospatial index. When using $geoNear,
the $geoNear pipeline operation must appear as the first stage in an aggregation pipeline.
For unsharded collections, when the aggregation pipeline only needs to access the indexed fields to fulfill its operations,
an index can cover (page 37) the pipeline.
Example
Consider the following index on the orders collection:
{ status: 1, amount: 1, cust_id: 1 }

This index can cover the following aggregation pipeline operation because MongoDB does not need to inspect the data
outside of the index to fulfill the operation:
db.orders.aggregate([
{ $match: { status: "A" } },
{ $group: { _id: "$cust_id", total: { $sum: "$amount" } } },
{ $sort: { total: -1 } }
])

6.2. Aggregation Concepts

283

MongoDB Documentation, Release 2.4.9

Early Filtering

If your aggregation operation requires only a subset of the data in a collection, use the $match, $limit, and $skip
stages to restrict the documents that enter at the beginning of the pipeline. When placed at the beginning of a pipeline,
$match operations use suitable indexes to scan only the matching documents in a collection.
Placing a $match pipeline stage followed by a $sort stage at the start of the pipeline is logically equivalent to a
single query with a sort and can use an index. When possible, place $match operators at the beginning of the pipeline.
Additional Features

The aggregation pipeline has an internal optimization phase that provides improved performance for certain sequences
of operators. For details, see Pipeline Sequence Optimization (page 289).
The aggregation pipeline supports operations on sharded collections. See Aggregation Pipeline and Sharded Collections (page 291).

6.2.2 Map-Reduce
Map-reduce is a data processing paradigm for condensing large volumes of data into useful aggregated results. For
map-reduce operations, MongoDB provides the mapReduce database command.
Consider the following map-reduce operation:
In this map-reduce operation, MongoDB applies the map phase to each input document (i.e. the documents in the
collection that match the query condition). The map function emits key-value pairs. For those keys that have multiple
values, MongoDB applies the reduce phase, which collects and condenses the aggregated data. MongoDB then stores
the results in a collection. Optionally, the output of the reduce function may pass through a finalize function to further
condense or process the results of the aggregation.
All map-reduce functions in MongoDB are JavaScript and run within the mongod process. Map-reduce operations
take the documents of a single collection as the input and can perform any arbitrary sorting and limiting before
beginning the map stage. mapReduce can return the results of a map-reduce operation as a document, or may write
the results to collections. The input and the output collections may be sharded.
Note: For most aggregation operations, the Aggregation Pipeline (page 281) provides better performance and more
coherent interface. However, map-reduce operations provide some flexibility that is not presently available in the
aggregation pipeline.

Map-Reduce JavaScript Functions

In MongoDB, map-reduce operations use custom JavaScript functions to map, or associate, values to a key. If a key
has multiple values mapped to it, the operation reduces the values for the key to a single object.
The use of custom JavaScript functions provide flexibility to map-reduce operations. For instance, when processing a
document, the map function can create more than one key and value mapping or no mapping. Map-reduce operations
can also use a custom JavaScript function to make final modifications to the results at the end of the map and reduce
operation, such as perform additional calculations.
Map-Reduce Behavior
In MongoDB, the map-reduce operation can write results to a collection or return the results inline. If you write
map-reduce output to a collection, you can perform subsequent map-reduce operations on the same input collection
284

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

Figure 6.5: Diagram of the annotated map-reduce operation.

6.2. Aggregation Concepts

285

MongoDB Documentation, Release 2.4.9

that merge replace, merge, or reduce new results with previous results. See mapReduce and Perform Incremental
Map-Reduce (page 302) for details and examples.
When returning the results of a map reduce operation inline, the result documents must be within the BSON
Document Size limit, which is currently 16 megabytes. For additional information on limits and restrictions on
map-reduce operations, see the http://docs.mongodb.org/manualreference/command/mapReduce
reference page.
MongoDB supports map-reduce operations on sharded collections (page 479). Map-reduce operations can also output
the results to a sharded collection. See Map-Reduce and Sharded Collections (page 291).

6.2.3 Single Purpose Aggregation Operations

Aggregation refers to a broad class of data manipulation operations that compute a result based on an input and a specific procedure. MongoDB provides a number of aggregation operations that perform specific aggregation operations
on a set of data.
Although limited in scope, particularly compared to the aggregation pipeline (page 281) and map-reduce (page 284),
these operations provide straightforward semantics for common data processing options.
Count
MongoDB can return a count of the number of documents that match a query. The count command as well as the
count() and cursor.count() methods provide access to counts in the mongo shell.
Example
Given a collection named records with only the following documents:
{
{
{
{

a:
a:
a:
a:

1,
1,
1,
2,

b:
b:
b:
b:

0
1
4
2

}
}
}
}

The following operation would count all documents in the collection and return the number 4:
db.records.count()

The following operation will count only the documents where the value of the field a is 1 and return 3:
db.records.count( { a: 1 } )

Distinct
The distinct operation takes a number of documents that match a query and returns all of the unique values for a field
in the matching documents. The distinct command and db.collection.distinct() method provide this
operation in the mongo shell. Consider the following examples of a distinct operation:
Example
Given a collection named records with only the following documents:
{
{
{
{

a:
a:
a:
a:

286

1,
1,
1,
1,

b:
b:
b:
b:

0
1
1
4

}
}
}
}

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

Figure 6.6: Diagram of the annotated distinct operation.

6.2. Aggregation Concepts

287

MongoDB Documentation, Release 2.4.9

{ a: 2, b: 2 }
{ a: 2, b: 2 }

Consider the following db.collection.distinct() operation which returns the distinct values of the field b:
db.records.distinct( "b" )

The results of this operation would resemble:

[ 0, 1, 4, 2 ]

Group
The group operation takes a number of documents that match a query, and then collects groups of documents based
on the value of a field or fields. It returns an array of documents with computed results for each group of documents.
Access the grouping functionality via the group command or the db.collection.group() method in the
mongo shell.
Warning: group does not support data in sharded collections. In addition, the results of the group operation
must be no larger than 16 megabytes.
Consider the following group operation:
Example
Given a collection named records with the following documents:
{
{
{
{
{
{
{

a:
a:
a:
a:
a:
a:
a:

1,
1,
1,
2,
2,
1,
4,

count:
count:
count:
count:
count:
count:
count:

4
2
4
3
1
5
4

}
}
}
}
}
}
}

Consider the following group operation which groups documents by the field a, where a is less than 3, and sums the
field count for each group:
db.records.group( {
key: { a: 1 },
cond: { a: { $lt: 3 } },
reduce: function(cur, result) { result.count += cur.count },
initial: { count: 0 }
} )

The results of this group operation would resemble the following:

[
{ a: 1, count: 15 },
{ a: 2, count: 4 }
]

See also:
The $group for related functionality in the aggregation pipeline (page 281).

288

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

6.2.4 Aggregation Mechanics

This section describes behaviors and limitations for the various aggregation modalities.
Aggregation Pipeline Optimization (page 289) Details the internal optimization of certain pipeline sequence.
Aggregation Pipeline Limits (page 290) Presents limitations on aggregation pipeline operations.
Aggregation Pipeline and Sharded Collections (page 291) Mechanics of aggregation pipeline operations on sharded
collections.
Map-Reduce and Sharded Collections (page 291) Mechanics of map-reduce operation with sharded collections.
Map Reduce Concurrency (page 292) Details the locks taken during map-reduce operations.
Aggregation Pipeline Optimization
Changed in version 2.4.
Aggregation pipeline operations have an optimization phase which attempts to rearrange the pipeline for improved
performance.
Pipeline Sequence Optimization

$sort + $skip + $limit Sequence Optimization When you have a sequence with $sort followed by a
$skip followed by a $limit, an optimization occurs that moves the $limit operator before the $skip operator.
For example, if the pipeline consists of the following stages:
{ $sort: { age : -1 } },
{ $skip: 10 },
{ $limit: 5 }

During the optimization phase, the optimizer transforms the sequence to the following:
{ $sort: { age : -1 } },
{ $limit: 15 }
{ $skip: 10 }

Note: The $limit value has increased to the sum of the initial value and the $skip value.
The optimized sequence now has $sort immediately preceding the $limit. See $sort for information on the
behavior of the $sort operation when it immediately precedes $limit.
$limit + $skip + $limit + $skip Sequence Optimization When you have a continuous sequence of a
$limit pipeline stage followed by a $skip pipeline stage, the optimization phase attempts to arrange the pipeline
stages to combine the limits and skips. For example, if the pipeline consists of the following stages:
{
{
{
{

$limit: 100 },
$skip: 5 },
$limit: 10},
$skip: 2 }

During the intermediate step, the optimizer reverses the position of the $skip followed by a $limit to $limit
followed by the $skip.

6.2. Aggregation Concepts

289

MongoDB Documentation, Release 2.4.9

{
{
{
{

$limit: 100 },
$limit: 15},
$skip: 5 },
$skip: 2 }

The $limit value has increased to the sum of the initial value and the $skip value. Then, for the final $limit
value, the optimizer selects the minimum between the adjacent $limit values. For the final $skip value, the
optimizer adds the adjacent $skip values, to transform the sequence to the following:
{ $limit: 15 },
{ $skip: 7 }

Projection Optimization

If the aggregation pipeline contains a $project stage that specifies the fields to include, then MongoDB applies the
projection to the head of the pipeline. This reduces the amount of data passing through the pipeline from the start.
In the following example, the $project stage specifies that the results of this stage return only the _id and the
amount fields. The optimization phase applies the projection to the head of the pipeline such that only the _id and
the amount fields return in the resulting documents from the $match stage as well.
db.orders.aggregate(
{ $match: { status: "A" } },
{ $project: { amount: 1 } }
)

Aggregation Pipeline Limits

Aggregation operations with the aggregate command have the following limitations.
Type Restrictions

The aggregation pipeline (page 281) cannot operate on values of the following types: Symbol, MinKey, MaxKey,
DBRef, Code, CodeWScope.
Changed in version 2.4: Removed restriction on Binary type data. In MongoDB 2.2, the pipeline could not operate
on Binary type data.
Result Size Restrictions

Output from the pipeline cannot exceed the BSON Document Size limit, which is currently 16 megabytes. If the
result set exceeds this limit, the aggregate command produces an error.
Memory Restrictions

If any single aggregation operation consumes more than 10 percent of system RAM, the operation will produce an
error.
Cumulative operators, such as $sort and $group, require access to the entire input set before they can produce any
output. These operators log a warning if the cumulative operator consumes 5% or more of the physical memory on the
host. Like any aggregation operation, these operators produce an error if they consume 10% or more of the physical
memory on the host. See the $sort and $group reference pages for details on their specific memory requirements
and use.
290

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

Aggregation Pipeline and Sharded Collections

The aggregation pipeline supports operations on sharded collections. This section describes behaviors specific to the
aggregation pipeline (page 281) and sharded collections.
Aggregation Operations in Sharded Clusters

When operating on a sharded collection, the aggregation pipeline is split into two parts. First, the aggregation pipeline
pushes all of the operators up to the first $group or $sort operation to each shard 1 . Then, a second pipeline runs
on the mongos. This pipeline consists of the first $group or $sort and any remaining pipeline operators, and runs
on the results received from the shards.
The $group operator brings in any sub-totals from the shards and combines them: in some cases these may be
structures. For example, the $avg expression maintains a total and count for each shard; mongos combines these
values and then divides.
Impact of Aggregation Pipelines on mongos

Changed in version 2.2.

Some aggregation pipeline operations will cause mongos instances to require more CPU resources than
in previous versions. This modified performance profile may dictate alternate architectural decisions if you use the
aggregation pipeline (page 281) extensively in a sharded environment.
Map-Reduce and Sharded Collections
Map-reduce supports operations on sharded collections, both as an input and as an output. This section describes the
behaviors of mapReduce specific to sharded collections.
Sharded Collection as Input

When using sharded collection as the input for a map-reduce operation, mongos will automatically dispatch the mapreduce job to each shard in parallel. There is no special option required. mongos will wait for jobs on all shards to
finish.
Sharded Collection as Output

Changed in version 2.2.

If the out field for mapReduce has the sharded value, MongoDB shards the output collection using the _id field
as the shard key.
To output to a sharded collection:
If the output collection does not exist, MongoDB creates and shards the collection on the _id field.
For a new or an empty sharded collection, MongoDB uses the results of the first stage of the map-reduce
operation to create the initial chunks distributed among the shards.
1

If an early $match can exclude shards through the use of the shard key in the predicate, then these operators are only pushed to the relevant
shards.

6.2. Aggregation Concepts

291

MongoDB Documentation, Release 2.4.9

mongos dispatches, in parallel, a map-reduce post-processing job to every shard that owns a chunk. During
the post-processing, each shard will pull the results for its own chunks from the other shards, run the final
reduce/finalize, and write locally to the output collection.
Note:
During later map-reduce jobs, MongoDB splits chunks as needed.
Balancing of chunks for the output collection is automatically prevented during post-processing to avoid concurrency issues.
In MongoDB 2.0:
mongos retrieves the results from each shard, performs a merge sort to order the results, and proceeds to the
reduce/finalize phase as needed. mongos then writes the result to the output collection in sharded mode.
This model requires only a small amount of memory, even for large data sets.
Shard chunks are not automatically split during insertion. This requires manual intervention until the chunks
are granular and balanced.
Important: For best results, only use the sharded output options for mapReduce in version 2.2 or later.

Map Reduce Concurrency

The map-reduce operation is composed of many tasks, including reads from the input collection, executions of the
map function, executions of the reduce function, writes to a temporary collection during processing, and writes to
the output collection.
During the operation, map-reduce takes the following locks:
The read phase takes a read lock. It yields every 100 documents.
The insert into the temporary collection takes a write lock for a single write.
If the output collection does not exist, the creation of the output collection takes a write lock.
If the output collection exists, then the output actions (i.e. merge, replace, reduce) take a write lock.
Changed in version 2.4: The V8 JavaScript engine, which became the default in 2.4, allows multiple JavaScript
operations to execute at the same time. Prior to 2.4, JavaScript code (i.e. map, reduce, finalize functions)
executed in a single thread.
Note: The final write lock during post-processing makes the results appear atomically. However, output actions
merge and reduce may take minutes to process. For the merge and reduce, the nonAtomic flag is available.
See the db.collection.mapReduce() reference for more information.

6.3 Aggregation Examples

This document provides the practical examples that display the capabilities of aggregation (page 281).
Aggregation with the Zip Code Data Set (page 293) Use the aggregation pipeline to group values and to calculate
aggregated sums and averages for a collection of United States zip codes.
Aggregation with User Preference Data (page 296) Use the pipeline to sort, normalize, and sum data on a collection
of user data.

292

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

Map-Reduce Examples (page 300) Define map-reduce operations that select ranges, group data, and calculate sums
and averages.
Perform Incremental Map-Reduce (page 302) Run a map-reduce operations over one collection and output results
to another collection.
Troubleshoot the Map Function (page 304) Steps to troubleshoot the map function.
Troubleshoot the Reduce Function (page 305) Steps to troubleshoot the reduce function.

6.3.1 Aggregation with the Zip Code Data Set

The examples in this document use the zipcode collection.
This collection is available at:
dia.mongodb.org/zips.json2 . Use mongoimport to load this data set into your mongod instance.

me-

Data Model
Each document in the zipcode collection has the following form:
{
"_id": "10280",
"city": "NEW YORK",
"state": "NY",
"pop": 5574,
"loc": [
-74.016323,
40.710537
]
}

The _id field holds the zip code as a string.

The city field holds the city.
The state field holds the two letter state abbreviation.
The pop field holds the population.
The loc field holds the location as a latitude longitude pair.
All of the following examples use the aggregate() helper in the mongo shell. aggregate() provides a wrapper
around the aggregate database command. See the documentation for your driver (page 95) for a more idiomatic
interface for data aggregation operations.
Return States with Populations above 10 Million
To return all states with a population greater than 10 million, use the following aggregation operation:
db.zipcodes.aggregate( { $group :
{ _id : "$state",
totalPop : { $sum : "$pop" } } },
{ $match : {totalPop : { $gte : 10*1000*1000 } } } )

Aggregations operations using the aggregate() helper process all documents in the zipcodes collection.
aggregate() connects a number of pipeline (page 281) operators, which define the aggregation process.
In this example, the pipeline passes all documents in the zipcodes collection through the following steps:
2 http://media.mongodb.org/zips.json

6.3. Aggregation Examples

293

MongoDB Documentation, Release 2.4.9

the $group operator collects all documents and creates documents for each state.
These new per-state documents have one field in addition to the _id field: totalPop which is a generated
field using the $sum operation to calculate the total value of all pop fields in the source documents.
After the $group operation the documents in the pipeline resemble the following:
{
"_id" : "AK",
"totalPop" : 550043
}

the $match operation filters these documents so that the only documents that remain are those where the value
of totalPop is greater than or equal to 10 million.
The $match operation does not alter the documents, which have the same format as the documents output by
$group.
The equivalent SQL for this operation is:
SELECT state, SUM(pop) AS totalPop
FROM zipcodes
GROUP BY state
HAVING totalPop >= (10*1000*1000)

Return Average City Population by State

To return the average populations for cities in each state, use the following aggregation operation:
db.zipcodes.aggregate( { $group :
{ _id : { state : "$state", city : "$city" },
pop : { $sum : "$pop" } } },
{ $group :
{ _id : "$_id.state",
avgCityPop : { $avg : "$pop" } } } )

Aggregations operations using the aggregate() helper process all documents in the zipcodes collection.
aggregate() connects a number of pipeline (page 281) operators that define the aggregation process.
In this example, the pipeline passes all documents in the zipcodes collection through the following steps:
the $group operator collects all documents and creates new documents for every combination of the city and
state fields in the source document.
After this stage in the pipeline, the documents resemble the following:
{
"_id" : {
"state" : "CO",
"city" : "EDGEWATER"
},
"pop" : 13154
}

the second $group operator collects documents by the state field and use the $avg expression to compute
a value for the avgCityPop field.
The final output of this aggregation operation is:

294

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

{
"_id" : "MN",
"avgCityPop" : 5335
},

Return Largest and Smallest Cities by State

To return the smallest and largest cities by population for each state, use the following aggregation operation:
db.zipcodes.aggregate( { $group:
{ _id: { state: "$state", city: "$city" },
pop: { $sum: "$pop" } } },
{ $sort: { pop: 1 } },
{ $group:
{ _id : "$_id.state",
biggestCity: { $last: "$_id.city" },
biggestPop:
{ $last: "$pop" },
smallestCity: { $first: "$_id.city" },
smallestPop: { $first: "$pop" } } },
// the following $project is optional, and
// modifies the output format.
{ $project:
{ _id: 0,
state: "$_id",
biggestCity: { name: "$biggestCity", pop: "$biggestPop" },
smallestCity: { name: "$smallestCity", pop: "$smallestPop" } } } )

Aggregation operations using the aggregate() helper process all documents in the zipcodes collection.
aggregate() combines a number of pipeline (page 281) operators that define the aggregation process.
All documents from the zipcodes collection pass into the pipeline, which consists of the following steps:
the $group operator collects all documents and creates new documents for every combination of the city and
state fields in the source documents.
By specifying the value of _id as a sub-document that contains both fields, the operation preserves the state
field for use later in the pipeline. The documents produced by this stage of the pipeline have a second field,
pop, which uses the $sum operator to provide the total of the pop fields in the source document.
At this stage in the pipeline, the documents resemble the following:
{
"_id" : {
"state" : "CO",
"city" : "EDGEWATER"
},
"pop" : 13154
}

$sort operator orders the documents in the pipeline based on the vale of the pop field from largest to smallest.
This operation does not alter the documents.
the second $group operator collects the documents in the pipeline by the state field, which is a field inside
the nested _id document.
Within each per-state document this $group operator specifies four fields: Using the $last expression, the
$group operator creates the biggestcity and biggestpop fields that store the city with the largest pop-

6.3. Aggregation Examples

295

MongoDB Documentation, Release 2.4.9

ulation and that population. Using the $first expression, the $group operator creates the smallestcity
and smallestpop fields that store the city with the smallest population and that population.
The documents, at this stage in the pipeline resemble the following:
{
"_id" : "WA",
"biggestCity" : "SEATTLE",
"biggestPop" : 520096,
"smallestCity" : "BENGE",
"smallestPop" : 2
}

The final operation is $project, which renames the _id field to state and moves the biggestCity,
biggestPop, smallestCity, and smallestPop into biggestCity and smallestCity subdocuments.
The output of this aggregation operation is:
{
"state" : "RI",
"biggestCity" : {
"name" : "CRANSTON",
"pop" : 176404
},
"smallestCity" : {
"name" : "CLAYVILLE",
"pop" : 45
}
}

6.3.2 Aggregation with User Preference Data

Data Model
Consider a hypothetical sports club with a database that contains a user collection that tracks the users join dates,
sport preferences, and stores these data in documents that resemble the following:
{
_id : "jane",
joined : ISODate("2011-03-02"),
likes : ["golf", "racquetball"]
}
{
_id : "joe",
joined : ISODate("2012-07-02"),
likes : ["tennis", "golf", "swimming"]
}

Normalize and Sort Documents

The following operation returns user names in upper case and in alphabetical order. The aggregation includes user
names for all documents in the users collection. You might do this to normalize user names for processing.
db.users.aggregate(
[
{ $project : { name:{$toUpper:"$_id"} , _id:0 } },

296

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

{ $sort : { name : 1 } }
]
)

All documents from the users collection pass through the pipeline, which consists of the following operations:
The $project operator:
creates a new field called name.
converts the value of the _id to upper case, with the $toUpper operator. Then the $project creates
a new field, named name to hold this value.
suppresses the id field. $project will pass the _id field by default, unless explicitly suppressed.
The $sort operator orders the results by the name field.
The results of the aggregation would resemble the following:
{
"name" : "JANE"
},
{
"name" : "JILL"
},
{
"name" : "JOE"
}

Return Usernames Ordered by Join Month

The following aggregation operation returns user names sorted by the month they joined. This kind of aggregation
could help generate membership renewal notices.
db.users.aggregate(
[
{ $project : { month_joined : {
$month : "$joined"
},
name : "$_id",
_id : 0
},
{ $sort : { month_joined : 1 } }
]
)

The pipeline passes all documents in the users collection through the following operations:
The $project operator:
Creates two new fields: month_joined and name.
Suppresses the id from the results. The aggregate() method includes the _id, unless explicitly
suppressed.
The $month operator converts the values of the joined field to integer representations of the month. Then
the $project operator assigns those values to the month_joined field.
The $sort operator sorts the results by the month_joined field.
The operation returns results that resemble the following:

6.3. Aggregation Examples

297

MongoDB Documentation, Release 2.4.9

{
"month_joined" : 1,
"name" : "ruth"
},
{
"month_joined" : 1,
"name" : "harold"
},
{
"month_joined" : 1,
"name" : "kate"
}
{
"month_joined" : 2,
"name" : "jill"
}

Return Total Number of Joins per Month

The following operation shows how many people joined each month of the year. You might use this aggregated data
for recruiting and marketing strategies.
db.users.aggregate(
[
{ $project : { month_joined : { $month : "$joined" } } } ,
{ $group : { _id : {month_joined:"$month_joined"} , number : { $sum : 1 } } },
{ $sort : { "_id.month_joined" : 1 } }
]
)

The pipeline passes all documents in the users collection through the following operations:
The $project operator creates a new field called month_joined.
The $month operator converts the values of the joined field to integer representations of the month. Then
the $project operator assigns the values to the month_joined field.
The $group operator collects all documents with a given month_joined value and counts how many documents there are for that value. Specifically, for each unique value, $group creates a new per-month document
with two fields:
_id, which contains a nested document with the month_joined field and its value.
number, which is a generated field. The $sum operator increments this field by 1 for every document
containing the given month_joined value.
The $sort operator sorts the documents created by $group according to the contents of the month_joined
field.
The result of this aggregation operation would resemble the following:
{
"_id" : {
"month_joined" : 1
},
"number" : 3
},
{
"_id" : {

298

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

"month_joined" : 2
},
"number" : 9
},
{
"_id" : {
"month_joined" : 3
},
"number" : 5
}

Return the Five Most Common Likes

The following aggregation collects top five most liked activities in the data set. This type of analysis could help
inform planning and future development.
db.users.aggregate(
[
{ $unwind : "$likes" },
{ $group : { _id : "$likes" , number : { $sum : 1 } } },
{ $sort : { number : -1 } },
{ $limit : 5 }
]
)

The pipeline begins with all documents in the users collection, and passes these documents through the following
operations:
The $unwind operator separates each value in the likes array, and creates a new version of the source
document for every element in the array.
Example
Given the following document from the users collection:
{
_id : "jane",
joined : ISODate("2011-03-02"),
likes : ["golf", "racquetball"]
}

The $unwind operator would create the following documents:

{
_id : "jane",
joined : ISODate("2011-03-02"),
likes : "golf"
}
{
_id : "jane",
joined : ISODate("2011-03-02"),
likes : "racquetball"
}

The $group operator collects all documents the same value for the likes field and counts each grouping.
With this information, $group creates a new document with two fields:
_id, which contains the likes value.
6.3. Aggregation Examples

299

MongoDB Documentation, Release 2.4.9

number, which is a generated field. The $sum operator increments this field by 1 for every document
containing the given likes value.
The $sort operator sorts these documents by the number field in reverse order.
The $limit operator only includes the first 5 result documents.
The results of aggregation would resemble the following:
{
"_id" : "golf",
"number" : 33
},
{
"_id" : "racquetball",
"number" : 31
},
{
"_id" : "swimming",
"number" : 24
},
{
"_id" : "handball",
"number" : 19
},
{
"_id" : "tennis",
"number" : 18
}

6.3.3 Map-Reduce Examples

In the mongo shell, the db.collection.mapReduce() method is a wrapper around the mapReduce command.
The following examples use the db.collection.mapReduce() method:
Consider the following map-reduce operations on a collection orders that contains documents of the following
prototype:
{
_id: ObjectId("50a8240b927d5d8b5891743c"),
cust_id: "abc123",
ord_date: new Date("Oct 04, 2012"),
status: 'A',
price: 25,
items: [ { sku: "mmm", qty: 5, price: 2.5 },
{ sku: "nnn", qty: 5, price: 2.5 } ]
}

Return the Total Price Per Customer

Perform the map-reduce operation on the orders collection to group by the cust_id, and calculate the sum of the
price for each cust_id:

1. Define the map function to process each input document:

In the function, this refers to the document that the map-reduce operation is processing.

300

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

The function maps the price to the cust_id for each document and emits the cust_id and price
pair.
var mapFunction1 = function() {
emit(this.cust_id, this.price);
};

2. Define the corresponding reduce function with two arguments keyCustId and valuesPrices:
The valuesPrices is an array whose elements are the price values emitted by the map function and
grouped by keyCustId.
The function reduces the valuesPrice array to the sum of its elements.
var reduceFunction1 = function(keyCustId, valuesPrices) {
return Array.sum(valuesPrices);
};

3. Perform the map-reduce on all documents in the orders collection using the mapFunction1 map function
and the reduceFunction1 reduce function.
db.orders.mapReduce(
mapFunction1,
reduceFunction1,
{ out: "map_reduce_example" }
)

This operation outputs the results to a collection named map_reduce_example.

If the
map_reduce_example collection already exists, the operation will replace the contents with the results of this map-reduce operation:
Calculate Order and Total Quantity with Average Quantity Per Item
In this example, you will perform a map-reduce operation on the orders collection for all documents that have
an ord_date value greater than 01/01/2012. The operation groups by the item.sku field, and calculates the
number of orders and the total quantity ordered for each sku. The operation concludes by calculating the average
quantity per order for each sku value:
1. Define the map function to process each input document:
In the function, this refers to the document that the map-reduce operation is processing.
For each item, the function associates the sku with a new object value that contains the count of 1
and the item qty for the order and emits the sku and value pair.
var mapFunction2 = function() {
for (var idx = 0; idx < this.items.length; idx++) {
var key = this.items[idx].sku;
var value = {
count: 1,
qty: this.items[idx].qty
};
emit(key, value);
}
};

2. Define the corresponding reduce function with two arguments keySKU and countObjVals:
countObjVals is an array whose elements are the objects mapped to the grouped keySKU values
passed by map function to the reducer function.

6.3. Aggregation Examples

301

MongoDB Documentation, Release 2.4.9

The function reduces the countObjVals array to a single object reducedValue that contains the
count and the qty fields.
In reducedVal, the count field contains the sum of the count fields from the individual array elements, and the qty field contains the sum of the qty fields from the individual array elements.
var reduceFunction2 = function(keySKU, countObjVals) {
reducedVal = { count: 0, qty: 0 };
for (var idx = 0; idx < countObjVals.length; idx++) {
reducedVal.count += countObjVals[idx].count;
reducedVal.qty += countObjVals[idx].qty;
}
return reducedVal;
};

3. Define a finalize function with two arguments key and reducedVal. The function modifies the
reducedVal object to add a computed field named avg and returns the modified object:
var finalizeFunction2 = function (key, reducedVal) {
reducedVal.avg = reducedVal.qty/reducedVal.count;
return reducedVal;
};

4. Perform the map-reduce operation on the orders collection

reduceFunction2, and finalizeFunction2 functions.

using

the

mapFunction2,

db.orders.mapReduce( mapFunction2,
reduceFunction2,
{
out: { merge: "map_reduce_example" },
query: { ord_date:
{ $gt: new Date('01/01/2012') }
},
finalize: finalizeFunction2
}
)

This operation uses the query field to select only those documents with ord_date greater than new
Date(01/01/2012). Then it output the results to a collection map_reduce_example. If the
map_reduce_example collection already exists, the operation will merge the existing contents with the
results of this map-reduce operation.

6.3.4 Perform Incremental Map-Reduce

Map-reduce operations can handle complex aggregation tasks. To perform map-reduce operations, MongoDB provides
the mapReduce command and, in the mongo shell, the db.collection.mapReduce() wrapper method.
If the map-reduce data set is constantly growing, you may want to perform an incremental map-reduce rather than
performing the map-reduce operation over the entire data set each time.
To perform incremental map-reduce:
1. Run a map-reduce job over the current collection and output the result to a separate collection.
2. When you have more data to process, run subsequent map-reduce job with:
302

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

the query parameter that specifies conditions that match only the new documents.
the out parameter that specifies the reduce action to merge the new results into the existing output
collection.
Consider the following example where you schedule a map-reduce operation on a sessions collection to run at the
end of each day.
Data Setup
The sessions collection contains documents that log users sessions each day, for example:
db.sessions.save(
db.sessions.save(
db.sessions.save(
db.sessions.save(

{
{
{
{

userid:
userid:
userid:
userid:

"a",
"b",
"c",
"d",

ts:
ts:
ts:
ts:

ISODate('2011-11-03
ISODate('2011-11-03
ISODate('2011-11-03
ISODate('2011-11-03

14:17:00'),
14:23:00'),
15:02:00'),
16:45:00'),

length:
length:
length:
length:

95 } );
110 } );
120 } );
45 } );

db.sessions.save(
db.sessions.save(
db.sessions.save(
db.sessions.save(

{
{
{
{

userid:
userid:
userid:
userid:

"a",
"b",
"c",
"d",

ts:
ts:
ts:
ts:

ISODate('2011-11-04
ISODate('2011-11-04
ISODate('2011-11-04
ISODate('2011-11-04

11:05:00'),
13:14:00'),
17:00:00'),
15:37:00'),

length:
length:
length:
length:

105 } );
120 } );
130 } );
65 } );

Initial Map-Reduce of Current Collection

Run the first map-reduce operation as follows:
1. Define the map function that maps the userid to an object that contains the fields userid, total_time,
count, and avg_time:
var mapFunction = function() {
var key = this.userid;
var value = {
userid: this.userid,
total_time: this.length,
count: 1,
avg_time: 0
};
emit( key, value );
};

2. Define the corresponding reduce function with two arguments key and values to calculate the total time and
the count. The key corresponds to the userid, and the values is an array whose elements corresponds to
the individual objects mapped to the userid in the mapFunction.
var reduceFunction = function(key, values) {
var reducedObject = {
userid: key,
total_time: 0,
count:0,
avg_time:0
};
values.forEach( function(value) {
reducedObject.total_time += value.total_time;
reducedObject.count += value.count;

6.3. Aggregation Examples

303

MongoDB Documentation, Release 2.4.9

}
);
return reducedObject;
};

3. Define the finalize function with two arguments key and reducedValue. The function modifies the
reducedValue document to add another field average and returns the modified document.
var finalizeFunction = function (key, reducedValue) {
if (reducedValue.count > 0)
reducedValue.avg_time = reducedValue.total_time / reducedValue.count;
return reducedValue;
};

4. Perform map-reduce on the session collection using the mapFunction, the reduceFunction, and the
finalizeFunction functions. Output the results to a collection session_stat. If the session_stat
collection already exists, the operation will replace the contents:
db.sessions.mapReduce( mapFunction,
reduceFunction,
{
out: { reduce: "session_stat" },
finalize: finalizeFunction
}
)

Subsequent Incremental Map-Reduce

Later, as the sessions collection grows, you can run additional map-reduce operations. For example, add new
documents to the sessions collection:
db.sessions.save(
db.sessions.save(
db.sessions.save(
db.sessions.save(

{
{
{
{

userid:
userid:
userid:
userid:

"a",
"b",
"c",
"d",

ts:
ts:
ts:
ts:

ISODate('2011-11-05
ISODate('2011-11-05
ISODate('2011-11-05
ISODate('2011-11-05

14:17:00'),
14:23:00'),
15:02:00'),
16:45:00'),

length:
length:
length:
length:

100 } );
115 } );
125 } );
55 } );

At the end of the day, perform incremental map-reduce on the sessions collection, but use the query field to select
only the new documents. Output the results to the collection session_stat, but reduce the contents with the
results of the incremental map-reduce:
db.sessions.mapReduce( mapFunction,
reduceFunction,
{
query: { ts: { $gt: ISODate('2011-11-05 00:00:00') } },
out: { reduce: "session_stat" },
finalize: finalizeFunction
}
);

6.3.5 Troubleshoot the Map Function

The map function is a JavaScript function that associates or maps a value with a key and emits the key and value
pair during a map-reduce (page 284) operation.

304

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

To verify the key and value pairs emitted by the map function, write your own emit function.
Consider a collection orders that contains documents of the following prototype:
{
_id: ObjectId("50a8240b927d5d8b5891743c"),
cust_id: "abc123",
ord_date: new Date("Oct 04, 2012"),
status: 'A',
price: 250,
items: [ { sku: "mmm", qty: 5, price: 2.5 },
{ sku: "nnn", qty: 5, price: 2.5 } ]
}

1. Define the map function that maps the price to the cust_id for each document and emits the cust_id and
price pair:
var map = function() {
emit(this.cust_id, this.price);
};

2. Define the emit function to print the key and value:

var emit = function(key, value) {
print("emit");
print("key: " + key + " value: " + tojson(value));
}

3. Invoke the map function with a single document from the orders collection:
var myDoc = db.orders.findOne( { _id: ObjectId("50a8240b927d5d8b5891743c") } );
map.apply(myDoc);

4. Verify the key and value pair is as you expected.

emit
key: abc123 value:250

5. Invoke the map function with multiple documents from the orders collection:
var myCursor = db.orders.find( { cust_id: "abc123" } );
while (myCursor.hasNext()) {
var doc = myCursor.next();
print ("document _id= " + tojson(doc._id));
map.apply(doc);
print();
}

6. Verify the key and value pairs are as you expected.

See also:
The map function must meet various requirements. For a list of all the requirements for the map function, see
mapReduce, or the mongo shell helper method db.collection.mapReduce().

6.3.6 Troubleshoot the Reduce Function

The reduce function is a JavaScript function that reduces to a single object all the values associated with a particular key during a map-reduce (page 284) operation. The reduce function must meet various requirements. This
tutorial helps verify that the reduce function meets the following criteria:
6.3. Aggregation Examples

305

MongoDB Documentation, Release 2.4.9

The reduce function must return an object whose type must be identical to the type of the value emitted by
the map function.
The order of the elements in the valuesArray should not affect the output of the reduce function.
The reduce function must be idempotent.
For a list of all the requirements for the reduce function, see mapReduce, or the mongo shell helper method
db.collection.mapReduce().
Confirm Output Type
You can test that the reduce function returns a value that is the same type as the value emitted from the map function.
1. Define a reduceFunction1 function that takes the arguments keyCustId and valuesPrices.
valuesPrices is an array of integers:
var reduceFunction1 = function(keyCustId, valuesPrices) {
return Array.sum(valuesPrices);
};

2. Define a sample array of integers:

var myTestValues = [ 5, 5, 10 ];

3. Invoke the reduceFunction1 with myTestValues:

reduceFunction1('myKey', myTestValues);

4. Verify the reduceFunction1 returned an integer:

20

5. Define a reduceFunction2 function that takes the arguments keySKU and valuesCountObjects.
valuesCountObjects is an array of documents that contain two fields count and qty:
var reduceFunction2 = function(keySKU, valuesCountObjects) {
reducedValue = { count: 0, qty: 0 };
for (var idx = 0; idx < valuesCountObjects.length; idx++) {
reducedValue.count += valuesCountObjects[idx].count;
reducedValue.qty += valuesCountObjects[idx].qty;
}
return reducedValue;
};

6. Define a sample array of documents:

var myTestObjects = [
{ count: 1, qty: 5 },
{ count: 2, qty: 10 },
{ count: 3, qty: 15 }
];

7. Invoke the reduceFunction2 with myTestObjects:

reduceFunction2('myKey', myTestObjects);

8. Verify the reduceFunction2 returned a document with exactly the count and the qty field:

306

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

{ "count" : 6, "qty" : 30 }

Ensure Insensitivity to the Order of Mapped Values

The reduce function takes a key and a values array as its argument. You can test that the result of the reduce
function does not depend on the order of the elements in the values array.
1. Define a sample values1 array and a sample values2 array that only differ in the order of the array elements:
var values1
{ count: 1,
{ count: 2,
{ count: 3,
];

= [
qty: 5 },
qty: 10 },
qty: 15 }

var values2
{ count: 3,
{ count: 1,
{ count: 2,
];

= [
qty: 15 },
qty: 5 },
qty: 10 }

2. Define a reduceFunction2 function that takes the arguments keySKU and valuesCountObjects.
valuesCountObjects is an array of documents that contain two fields count and qty:
var reduceFunction2 = function(keySKU, valuesCountObjects) {
reducedValue = { count: 0, qty: 0 };
for (var idx = 0; idx < valuesCountObjects.length; idx++) {
reducedValue.count += valuesCountObjects[idx].count;
reducedValue.qty += valuesCountObjects[idx].qty;
}
return reducedValue;
};

3. Invoke the reduceFunction2 first with values1 and then with values2:
reduceFunction2('myKey', values1);
reduceFunction2('myKey', values2);

4. Verify the reduceFunction2 returned the same result:

{ "count" : 6, "qty" : 30 }

Ensure Reduce Function Idempotence

Because the map-reduce operation may call a reduce multiple times for the same key, and wont call a reduce for
single instances of a key in the working set, the reduce function must return a value of the same type as the value
emitted from the map function. You can test that the reduce function process reduced values without affecting the
final value.
1. Define a reduceFunction2 function that takes the arguments keySKU and valuesCountObjects.
valuesCountObjects is an array of documents that contain two fields count and qty:
var reduceFunction2 = function(keySKU, valuesCountObjects) {
reducedValue = { count: 0, qty: 0 };

6.3. Aggregation Examples

307

MongoDB Documentation, Release 2.4.9

for (var idx = 0; idx < valuesCountObjects.length; idx++) {

reducedValue.count += valuesCountObjects[idx].count;
reducedValue.qty += valuesCountObjects[idx].qty;
}
return reducedValue;
};

2. Define a sample key:

var myKey = 'myKey';

3. Define a sample valuesIdempotent array that contains an element that is a call to the reduceFunction2
function:
var valuesIdempotent = [
{ count: 1, qty: 5 },
{ count: 2, qty: 10 },
reduceFunction2(myKey, [ { count:3, qty: 15 } ] )
];

4. Define a sample values1 array that combines the values passed to reduceFunction2:
var values1
{ count: 1,
{ count: 2,
{ count: 3,
];

= [
qty: 5 },
qty: 10 },
qty: 15 }

5. Invoke the reduceFunction2 first with myKey and valuesIdempotent and then with myKey and
values1:
reduceFunction2(myKey, valuesIdempotent);
reduceFunction2(myKey, values1);

6. Verify the reduceFunction2 returned the same result:

{ "count" : 6, "qty" : 30 }

6.4 Aggregation Reference

Aggregation Commands Comparison (page 308) A comparison of group, mapReduce and aggregate that explores the strengths and limitations of each aggregation modality.
http://docs.mongodb.org/manualreference/operator/aggregation Aggregation pipeline operations have a collection of operators available to define and manipulate documents in pipeline stages.
SQL to Aggregation Mapping Chart (page 310) An overview common aggregation operations in SQL and MongoDB using the aggregation pipeline and operators in MongoDB and common SQL statements.
Aggregation Interfaces (page 312) The data aggregation interfaces document the invocation format and output for
MongoDBs aggregation commands and methods.

6.4.1 Aggregation Commands Comparison

The following table provides a brief overview of the features of the MongoDB aggregation commands.

308

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

aggregate
De- New in version 2.2.
scrip- Designed with specific goals of
tion improving performance and
usability for aggregation tasks.
Uses a pipeline approach
where objects are transformed as
they pass through a series of
pipeline operators such as
$group, $match, and $sort.
See Aggregation Reference
(page 308) for more information
on the pipeline operators.
Key Pipeline operators can be
Fea- repeated as needed.
tures Pipeline operators need not
produce one output document for
every input document.
Can also generate new
documents or filter out
documents.

Flexibility

Limited to the operators and

expressions supported by the
aggregation pipeline.
However, can add computed
fields, create new virtual
sub-objects, and extract
sub-fields into the top-level of
results by using the $project
pipeline operator.
See $project for more
information as well as
Aggregation Reference
(page 308) for more information
on all the available pipeline
operators.
Out- Returns results inline.
put The result is subject to the BSON
Re- Document size limit.
sults

Shard-Supports non-sharded and

ing sharded input collections.
Notes
More See Aggregation Concepts
In(page 281) and aggregate.
for6.4.
ma-Aggregation Reference
tion

mapReduce
Implements the Map-Reduce
aggregation for processing large
data sets.

group
Provides grouping functionality.
Is slower than the aggregate
command and has less
functionality than the
mapReduce command.

In addition to grouping
operations, can perform complex
aggregation tasks as well as
perform incremental aggregation
on continuously growing
datasets.
See Map-Reduce Examples
(page 300) and Perform
Incremental Map-Reduce
(page 302).
Custom map, reduce and
finalize JavaScript functions
offer flexibility to aggregation
logic.
See mapReduce for details and
restrictions on the functions.

Can either group by existing

fields or with a custom keyf
JavaScript function, can group by
calculated fields.
See group for information and
example using the keyf
function.

Returns results in various options

(inline, new collection, merge,
replace, reduce). See
mapReduce for details on the
output options.
Changed in version 2.2: Provides
much better support for sharded
map-reduce output than previous
versions.

Returns results inline as an array

of grouped items.
The result set must fit within the
maximum BSON document size
limit.
Changed in version 2.2: The
returned array can contain at
most 20,000 elements; i.e. at
most 20,000 unique groupings.
Previous versions had a limit of
10,000 elements.
Does not support sharded
collection.
Prior to 2.4, JavaScript code
executed in a single thread.
See group.

Supports non-sharded and

sharded input collections.
Prior to 2.4, JavaScript code
executed in a single thread.
See Map-Reduce (page 284) and
mapReduce.

Custom reduce and

finalize JavaScript functions
offer flexibility to grouping logic.
See group for details and
restrictions on these functions.

309

MongoDB Documentation, Release 2.4.9

6.4.2 SQL to Aggregation Mapping Chart

The aggregation pipeline (page 281) allows MongoDB to provide native aggregation capabilities that corresponds
to many common data aggregation operations in SQL. If youre new to MongoDB you might want to consider the
Frequently Asked Questions (page 555) section for a selection of common questions.
The following table provides an overview of common SQL aggregation terms, functions, and concepts and the corresponding MongoDB aggregation operators:
SQL Terms,
Functions, and
Concepts
WHERE
GROUP BY
HAVING
SELECT
ORDER BY
LIMIT
SUM()
COUNT()
join

MongoDB Aggregation Operators

$match
$group
$match
$project
$sort
$limit
$sum
$sum
No direct corresponding operator; however, the $unwind operator allows for
somewhat similar functionality, but with fields embedded within the document.

Examples
The following table presents a quick reference of SQL aggregation statements and the corresponding MongoDB statements. The examples in the table assume the following conditions:
The SQL examples assume two tables, orders and order_lineitem that join by the
order_lineitem.order_id and the orders.id columns.
The MongoDB examples assume one collection orders that contain documents of the following prototype:
{
cust_id: "abc123",
ord_date: ISODate("2012-11-02T17:04:11.102Z"),
status: 'A',
price: 50,
items: [ { sku: "xxx", qty: 25, price: 1 },
{ sku: "yyy", qty: 25, price: 1 } ]
}

The MongoDB statements prefix the names of the fields from the documents in the collection orders with a $
character when they appear as operands to the aggregation operations.

310

Chapter 6. Aggregation

MongoDB Documentation, Release 2.4.9

SQL Example

MongoDB Example

SELECT COUNT(*) AS count

FROM orders

db.orders.aggregate( [
{ $group: { _id: null,
count: { $sum: 1 } } }
] )

Description
Count all records from orders

Sum the price field from orders

SELECT SUM(price) AS total db.orders.aggregate( [
FROM orders
{ $group: { _id: null,
total: { $sum: "$price" } } }
] )
SELECT cust_id,
SUM(price) AS total
FROM orders
GROUP BY cust_id

For each unique cust_id, sum the

db.orders.aggregate( [
price field.
{ $group: { _id: "$cust_id",
total: { $sum: "$price" } } }
] )

SELECT cust_id,
SUM(price) AS total
FROM orders
GROUP BY cust_id
ORDER BY total

For each unique cust_id, sum the

db.orders.aggregate( [
price field, results sorted by sum.
{ $group: { _id: "$cust_id",
total: { $sum: "$price" } } },
{ $sort: { total: 1 } }
] )

SELECT cust_id,
ord_date,
SUM(price) AS total
FROM orders
GROUP BY cust_id, ord_date

For
each
unique
cust_id,
db.orders.aggregate( [
ord_date grouping, sum the
{ $group: { _id: { cust_id: "$cust_id",
price field.
ord_date: "$ord_date" },
total: { $sum: "$price" } } }
] )

SELECT cust_id, count(*)

FROM orders
GROUP BY cust_id
HAVING count(*) > 1

For cust_id with multiple records,

db.orders.aggregate( [
return the cust_id and the corre{ $group: { _id: "$cust_id",
sponding record count.
count: { $sum: 1 } } },
{ $match: { count: { $gt: 1 } } }
] )

SELECT cust_id,
ord_date,
SUM(price) AS total
FROM orders
GROUP BY cust_id, ord_date
HAVING total > 250

For
each
unique
cust_id,
db.orders.aggregate( [
ord_date grouping, sum the
{ $group: { _id: { cust_id: "$cust_id",
price field and return only where
ord_date: "$ord_date" },
the sum is greater than 250.
total: { $sum: "$price" } } },
{ $match: { total: { $gt: 250 } } }
] )

SELECT cust_id,
SUM(price) as total
FROM orders
WHERE status = 'A'
GROUP BY cust_id

For each unique cust_id with stadb.orders.aggregate( [

tus A, sum the price field.
{ $match: { status: 'A' } },
{ $group: { _id: "$cust_id",
total: { $sum: "$price" } } }
] )

SELECT cust_id,
SUM(price) as total
FROM orders
WHERE status = 'A'
GROUP
BY cust_id
6.4.
Aggregation
Reference
HAVING total > 250

For each unique cust_id with stadb.orders.aggregate( [

tus A, sum the price field and return
{ $match: { status: 'A' } },
only where the sum is greater than
{ $group: { _id: "$cust_id",
250.
total: { $sum: "$price" } } },
{ $match: { total: { $gt: 250 } } }
311
] )

SELECT cust_id,

db.orders.aggregate( [

For each unique cust_id, sum the

MongoDB Documentation, Release 2.4.9

6.4.3 Aggregation Interfaces

Aggregation Commands
Name
aggregate
count
distinct
group
mapReduce

Description
Performs aggregation tasks (page 281) such as group using the aggregation framework.
Counts the number of documents in a collection.
Displays the distinct values found for a specified key in a collection.
Groups documents in a collection by the specified key and performs simple aggregation.
Performs map-reduce (page 284) aggregation for large data sets.

Aggregation Methods
Name
Description
db.collection.aggregate()Provides access to the aggregation pipeline (page 281).
db.collection.group()
Groups documents in a collection by the specified key and performs simple
aggregation.
db.collection.mapReduce()Performs map-reduce (page 284) aggregation for large data sets.

312

Chapter 6. Aggregation

CHAPTER 7

Indexes

Indexes provide high performance read operations for frequently used queries.
This section introduces indexes in MongoDB, describes the types and configuration options for indexes, and describes
special types of indexing MongoDB supports. The section also provides tutorials detailing procedures and operational
concerns, and providing information on how applications may use indexes.
Index Introduction (page 313) An introduction to indexes in MongoDB.
Index Concepts (page 318) The core documentation of indexes in MongoDB, including geospatial and text indexes.
Index Types (page 319) MongoDB provides different types of indexes for different purposes and different types
of content.
Index Properties (page 334) The properties you can specify when building indexes.
Index Creation (page 336) The options available when creating indexes.
Indexing Tutorials (page 339) Examples of operations involving indexes, including index creation and querying indexes.
Indexing Reference (page 375) Reference material for indexes in MongoDB.

7.1 Index Introduction

Indexes support the efficient execution of queries in MongoDB. Without indexes MongoDB must scan every document
in a collection to select those documents that match the query statement. These collection scans are inefficient because
they require mongod to process a larger volume of data than an index for each operation.
Indexes are special data structures 1 that store a small portion of the collections data set in an easy to traverse form.
The index stores the value of a specific field or set of fields, ordered by the value of the field.
Fundamentally, indexes in MongoDB are similar to indexes in other database systems. MongoDB defines indexes at
the collection level and supports indexes on any field or sub-field of the documents in a MongoDB collection.
If an appropriate index exists for a query, MongoDB can use the index to limit the number of documents it must
inspect. In some cases, MongoDB can use the data from the index to determine which documents match a query. The
following diagram illustrates a query that selects documents using an index.
Consider the documentation of the query optimizer (page 37) for more information on the relationship between queries
and indexes.
Tip
1

MongoDB indexes use a B-tree data structure.

313

MongoDB Documentation, Release 2.4.9

Figure 7.1: Diagram of a query selecting documents using an index. MongoDB narrows the query by scanning the
range of documents with values of score less than 30.
Create indexes to support common and user-facing queries. Having these indexes will ensure that MongoDB only
scans the smallest possible number of documents.
Indexes can also optimize the performance of other operations in specific situations:
Sorted Results
MongoDB can use indexes to return documents sorted by the index key directly from the index without
requiring an additional sort phase.
Covered Results
When the query criteria and the projection of a query include only the indexed fields, MongoDB will return
results directly from the index without scanning any documents or bringing documents into memory.
These covered queries can be very efficient. Indexes can also cover aggregation pipeline operations
(page 281).

7.1.1 Index Types

MongoDB provides an umber of different index types to support specific types of data and queries.
Default _id
All MongoDB collections have an index on the _id field that exists by default. If applications do not specify a value
for _id the driver or the mongod will create an _id field with an ObjectId value.
The _id index is unique, and prevents clients from inserting two documents with the same value for the _id field.

314

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

Figure 7.2: Diagram of a query that uses an index to select and return sorted results. The index stores score values
in ascending order. MongoDB can traverse the index in either ascending or descending order to return sorted results.

Figure 7.3: Diagram of a query that uses only the index to match the query criteria and return the results. MongoDB
does not need to inspect data outside of the index to fulfill the query.

7.1. Index Introduction

315

MongoDB Documentation, Release 2.4.9

Single Field
In addition to the MongoDB-defined _id index, MongoDB supports user-defined indexes on a single field of a document (page 320). Consider the following illustration of a single-field index:

Figure 7.4: Diagram of an index on the score field (ascending).

Compound Index
MongoDB also supports user-defined indexes on multiple fields. These compound indexes (page 322) behave like
single-field indexes; however, the query can select documents based on additional fields. The order of fields listed
in a compound index has significance. For instance, if a compound index consists of { userid: 1, score:
-1 }, the index sorts first by userid and then, within each userid value, sort by score. Consider the following
illustration of this compound index:

Figure 7.5: Diagram of a compound index on the userid field (ascending) and the score field (descending). The
index sorts first by the userid field and then by the score field.

316

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

Multikey Index
MongoDB uses multikey indexes (page 324) to index the content stored in arrays. If you index a field that holds an
array value, MongoDB creates separate index entries for every element of the array. These multikey indexes (page 324)
allow queries to select documents that contain arrays by matching on element or elements of the arrays. MongoDB
automatically determines whether to create a multikey index if the indexed field contains an array value; you do not
need to explicitly specify the multikey type.
Consider the following illustration of a multikey index:

Figure 7.6: Diagram of a multikey index on the addr.zip field. The addr field contains an array of address
documents. The address documents contain the zip field.

Geospatial Index
To support efficient queries of geospatial coordinate data, MongoDB provides two special indexes: 2d indexes
(page 329) that uses planar geometry when returning results and 2sphere indexes (page 328) that use spherical geometry to return results.
See 2d Index Internals (page 331) for a high level introduction to geospatial indexes.
Text Indexes
MongoDB provides a beta text index type that supports searching for string content in a collection. These text
indexes do not store language-specific stop words (e.g. the, a, or) and stem the words in a collection to only
store root words.
See Text Indexes (page 332) for more information on text indexes and search.

7.1. Index Introduction

317

MongoDB Documentation, Release 2.4.9

Hashed Indexes
To support hash based sharding (page 493), MongoDB provides a hashed index (page 333) type, which indexes the
hash of the value of a field. These indexes have a more random distribution of values along their range, but only
support equality matches and cannot support range-based queries.

7.1.2 Index Properties

Unique Indexes
The unique (page 334) property for an index causes MongoDB to reject duplicate values for the indexed field. To
create a unique index (page 334) on a field that already has duplicate values, see Drop Duplicates (page 338) for
index creation options. Other than the unique constraint, unique indexes are functionally interchangeable with other
MongoDB indexes.
Sparse Indexes
The sparse (page 335) property of an index ensures that the index only contain entries for documents that have the
indexed field. The index skips documents that do not have the indexed field.
You can combine the sparse index option with the unique index option to reject documents that have duplicate values
for a field but ignore documents that do not have the indexed key.

7.2 Index Concepts

These documents describe and provide examples of the types, configuration options, and behavior of indexes in MongoDB. For an over view of indexing, see Index Introduction (page 313). For operational instructions, see Indexing
Tutorials (page 339). The Indexing Reference (page 375) documents the commands and operations specific to index
construction, maintenance, and querying in MongoDB, including index types and creation options.
Index Types (page 319) MongoDB provides different types of indexes for different purposes and different types of
content.
Single Field Indexes (page 320) A single field index only includes data from a single field of the documents in
a collection. MongoDB supports single field indexes on fields at the top level of a document and on fields
in sub-documents.
Compound Indexes (page 322) A compound index includes more than one field of the documents in a collection.
Multikey Indexes (page 324) A multikey index references an array and records a match if a query includes any
value in the array.
Geospatial Indexes and Queries (page 326) Geospatial indexes support location-based searches on data that is
stored as either GeoJSON objects or legacy coordinate pairs.
Text Indexes (page 332) Text indexes supports search of string content in documents.
Hashed Index (page 333) Hashed indexes maintain entries with hashes of the values of the indexed field.
Index Properties (page 334) The properties you can specify when building indexes.
TTL Indexes (page 334) The TTL index is used for TTL collections, which expire data after a period of time.
Unique Indexes (page 334) A unique index causes MongoDB to reject all documents that contain a duplicate
value for the indexed field.

318

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

Sparse Indexes (page 335) A sparse index does not index documents that do not have the indexed field.
Index Creation (page 336) The options available when creating indexes.

7.2.1 Index Types

MongoDB provides a number of different index types. You can create indexes on any field or embedded field within
a document or sub-document. You can create single field indexes (page 320) or compound indexes (page 322). MongoDB also supports indexes of arrays, called multi-key indexes (page 324), as well as supports indexes on geospatial
data (page 326). For a list of the supported index types, see Index Type Documentation (page 319).
In general, you should create indexes that support your common and user-facing queries. Having these indexes will
ensure that MongoDB scans the smallest possible number of documents.
In the mongo shell, you can create an index by calling the ensureIndex() method. For more detailed instructions
about building indexes, see the Indexing Tutorials (page 339) page.
Behavior of Index Types
All indexes in MongoDB are B-tree indexes, which can efficiently support equality matches and range queries. The
index stores items internally in order sorted by the value of the index field. The ordering of index entries supports
efficient range-based operations and allows MongoDB to return sorted results using the order of documents in the
index.
Ordering of Indexes

MongoDB indexes may be ascending, (i.e. 1) or descending (i.e. -1) in their ordering. Nevertheless, MongoDB may
also traverse the index in either directions. As a result, for single-field indexes, ascending and descending indexes are
interchangeable. This is not the case for compound indexes: in compound indexes, the direction of the sort order can
have a greater impact on the results.
See Sort Order (page 323) for more information on the impact of index order on results in compound indexes.
Redundant Indexes

A single query can only use one index, except for queries that use the $or operator that can use a different index for
each clause.
See also:
Index Limitations.
Index Type Documentation
Single Field Indexes (page 320) A single field index only includes data from a single field of the documents in a
collection. MongoDB supports single field indexes on fields at the top level of a document and on fields in
sub-documents.
Compound Indexes (page 322) A compound index includes more than one field of the documents in a collection.
Multikey Indexes (page 324) A multikey index references an array and records a match if a query includes any value
in the array.

7.2. Index Concepts

319

MongoDB Documentation, Release 2.4.9

Geospatial Indexes and Queries (page 326) Geospatial indexes support location-based searches on data that is stored
as either GeoJSON objects or legacy coordinate pairs.
Text Indexes (page 332) Text indexes supports search of string content in documents.
Hashed Index (page 333) Hashed indexes maintain entries with hashes of the values of the indexed field.
Single Field Indexes

MongoDB provides complete support for indexes on any field in a collection of documents. By default, all collections
have an index on the _id field (page 321), and applications and users may add additional indexes to support important
queries and operations.
MongoDB supports indexes that contain either a single field or multiple fields depending on the operations that index
supports. This document describes indexes that contain a single field. Consider the following illustration of a single
field index.

Figure 7.7: Diagram of an index on the score field (ascending).

See also:
Compound Indexes (page 322) for information about indexes that include multiple fields, and Index Introduction
(page 313) for a higher level introduction to indexing in MongoDB.
Example Given the following document in the friends collection:
{ "_id" : ObjectId(...),
"name" : "Alice"
"age" : 27
}

The following command creates an index on the name field:

db.friends.ensureIndex( { "name" : 1 } )

Cases

320

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

_id Field Index MongoDB creates the _id index, which is an ascending unique index (page 334) on the _id field
for all collections when the collection is created. You cannot remove the index on the _id field.
Think of the _id field as the primary key for a collection. Every document must have a unique _id field. You may
store any unique value in the _id field. The default value of _id is an ObjectId on every insert() operation. An
ObjectId is a 12-byte unique identifiers suitable for use as the value of an _id field.
Note: In sharded clusters, if you do not use the _id field as the shard key, then your application must ensure the
uniqueness of the values in the _id field to prevent errors. This is most-often done by using a standard auto-generated
ObjectId.
Before version 2.2, capped collections did not have an _id field. In version 2.2 and newer, capped collection do
have an _id field, except those in the local database. See Capped Collections Recommendations and Restrictions
(page 161) for more information.

Indexes on Embedded Fields You can create indexes on fields embedded in sub-documents, just as you can index
top-level fields in documents. Indexes on embedded fields differ from indexes on sub-documents (page 321), which
include the full content up to the maximum index size of the sub-document in the index. Instead, indexes on
embedded fields allow you to use a dot notation, to introspect into sub-documents.
Consider a collection named people that holds documents that resemble the following example document:
{"_id": ObjectId(...)
"name": "John Doe"
"address": {
"street": "Main"
"zipcode": 53511
"state": "WI"
}
}

You can create an index on the address.zipcode field, using the following specification:
db.people.ensureIndex( { "address.zipcode": 1 } )

Indexes on Subdocuments You can also create indexes on subdocuments.

For example, the factories collection contains documents that contain a metro field, such as:
{
_id: ObjectId("523cba3c73a8049bcdbf6007"),
metro: {
city: "New York",
state: "NY"
},
name: "Giant Factory"
}

The metro field is a subdocument, containing the embedded fields city and state. The following creates an index
on the metro field as a whole:
db.factories.ensureIndex( { metro: 1 } )

The following query can use the index on the metro field:
db.factories.find( { metro: { city: "New York", state: "NY" } } )

7.2. Index Concepts

321

MongoDB Documentation, Release 2.4.9

This query returns the above document. When performing equality matches on subdocuments, field order matters and
the subdocuments must match exactly. For example, the following query does not match the above document:
db.factories.find( { metro: { state: "NY", city: "New York" } } )

See query-subdocuments for more information regarding querying on subdocuments.

Compound Indexes

MongoDB supports compound indexes, where a single index structure holds references to multiple fields
collections documents. The following diagram illustrates an example of a compound index on two fields:

within a

Figure 7.8: Diagram of a compound index on the userid field (ascending) and the score field (descending). The
index sorts first by the userid field and then by the score field.
Compound indexes can support queries that match on multiple fields.
Example
Consider a collection named products that holds documents that resemble the following document:
{
"_id": ObjectId(...)
"item": "Banana"
"category": ["food", "produce", "grocery"]
"location": "4th Street Store"
"stock": 4
"type": cases
"arrival": Date(...)
}

If applications query on the item field as well as query on both the item field and the stock field, you can specify
a single compound index to support both of these queries:
db.products.ensureIndex( { "item": 1, "stock": 1 } )

Important: You may not create compound indexes that have hashed index fields. You will receive an error if you
attempt to create a compound index that includes a hashed index (page 333).
2

MongoDB imposes a limit of 31 fields for any compound index.

322

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

The order of the fields in a compound index is very important. In the previous example, the index will contain
references to documents sorted first by the values of the item field and, within each value of the item field, sorted
by values of the stock field. See Sort Order (page 323) for more information.
In addition to supporting queries that match on all the index fields, compound indexes can support queries that match
on the prefix of the index fields. For details, see Prefixes (page 323).
Sort Order Indexes store references to fields in either ascending (1) or descending (-1) sort order. For single-field
indexes, the sort order of keys doesnt matter because MongoDB can traverse the index in either direction. However,
for compound indexes (page 322), sort order can matter in determining whether the index can support a sort operation.
Consider a collection events that contains documents with the fields username and date. Applications can issue
queries that return results sorted first by ascending username values and then by descending (i.e. more recent to last)
date values, such as:
db.events.find().sort( { username: 1, date: -1 } )

or queries that return results sorted first by descending username values and then by ascending date values, such
as:
db.events.find().sort( { username: -1, date: 1 } )

The following index can support both these sort operations:

db.events.ensureIndex( { "username" : 1, "date" : -1 } )

However, the above index cannot support sorting by ascending username values and then by ascending date values,
such as the following:
db.events.find().sort( { username: 1, date: 1 } )

Prefixes Compound indexes support queries on any prefix of the index fields. Index prefixes are the beginning
subset of indexed fields. For example, given the index { a: 1, b: 1, c: 1 }, both { a: 1 } and {
a: 1, b: 1 } are prefixes of the index.
If you have a collection that has a compound index on { a: 1, b: 1 }, as well as an index that consists of the
prefix of that index, i.e. { a: 1 }, assuming none of the index has a sparse or unique constraints, then you can
drop the { a: 1 } index. MongoDB will be able to use the compound index in all of situations that it would have
used the { a: 1 } index.
Example
Given the following index:
{ "item": 1, "location": 1, "stock": 1 }

MongoDB can use this index to support queries that include:

the item field,
the item field and the location field,
the item field and the location field and the stock field, or
only the item and stock fields; however, this index would be less efficient than an index on only item and
stock.
MongoDB cannot use this index to support queries that include:

7.2. Index Concepts

323

MongoDB Documentation, Release 2.4.9

only the location field,

only the stock field, or
only the location and stock fields.

Multikey Indexes

To index a field that holds an array value, MongoDB adds index items for each item in the array. These multikey indexes
allow MongoDB to return documents from queries using the value of an array. MongoDB automatically determines
whether to create a multikey index if the indexed field contains an array value; you do not need to explicitly specify
the multikey type.
Consider the following illustration of a multikey index:

Figure 7.9: Diagram of a multikey index on the addr.zip field. The addr field contains an array of address
documents. The address documents contain the zip field.
Multikey indexes support all operations supported by other MongoDB indexes; however, applications may use multikey indexes to select documents based on ranges of values for the value of an array. Multikey indexes support arrays
that hold both values (e.g. strings, numbers) and nested documents.
Limitations
Interactions between Compound and Multikey Indexes While you can create multikey compound indexes
(page 322), at most one field in a compound index may hold an array. For example, given an index on { a: 1,
b: 1 }, the following documents are permissible:

324

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

{a: [1, 2], b: 1}

{a: 1, b: [1, 2]}

However, the following document is impermissible, and MongoDB cannot insert such a document into a collection
with the {a: 1, b: 1 } index:
{a: [1, 2], b: [1, 2]}

If you attempt to insert a such a document, MongoDB will reject the insertion, and produce an error that says cannot
index parallel arrays. MongoDB does not index parallel arrays because they require the index to include
each value in the Cartesian product of the compound keys, which could quickly result in incredibly large and difficult
to maintain indexes.
Shard Keys
Important: The index of a shard key cannot be a multi-key index.

Hashed Indexes hashed indexes are not compatible with multi-key indexes.
To compute the hash for a hashed index, MongoDB collapses sub-documents and computes the hash for the entire
value. For fields that hold arrays or sub-documents, you cannot use the index to support queries that introspect the
sub-document.
Examples
Index Basic Arrays Given the following document:
{
"_id" : ObjectId("..."),
"name" : "Warm Weather",
"author" : "Steve",
"tags" : [ "weather", "hot", "record", "april" ]
}

Then an index on the tags field, { tags:

entries for that document:

1 }, would be a multikey index and would include these four separate

"weather",
"hot",
"record", and
"april".
Queries could use the multikey index to return queries for any of the above values.
Index Arrays with Embedded Documents You can create multikey indexes on fields in objects embedded in arrays,
as in the following example:
Consider a feedback collection with documents in the following form:
{
"_id": ObjectId(...),
"title": "Grocery Quality",
"comments": [

7.2. Index Concepts

325

MongoDB Documentation, Release 2.4.9

{ author_id: ObjectId(...),
date: Date(...),
text: "Please expand the cheddar selection." },
{ author_id: ObjectId(...),
date: Date(...),
text: "Please expand the mustard selection." },
{ author_id: ObjectId(...),
date: Date(...),
text: "Please expand the olive selection." }
]
}

An index on the comments.text field would be a multikey index and would add items to the index for all embedded
documents in the array.
With the index { "comments.text":

1 } on the feedback collection, consider the following query:

db.feedback.find( { "comments.text": "Please expand the olive selection." } )

The query would select the documents in the collection that contain the following embedded document in the
comments array:
{ author_id: ObjectId(...),
date: Date(...),
text: "Please expand the olive selection." }

Geospatial Indexes and Queries

MongoDB offers a number of indexes and query mechanisms to handle geospatial information. This section introduces
MongoDBs geospatial features. For complete examples of geospatial queries in MongoDB, see Geospatial Index
Tutorials (page 350).
Surfaces Before storing your location data and writing queries, you must decide the type of surface to use to perform
calculations. The type you choose affects how you store data, what type of index to build, and the syntax of your
queries.
MongoDB offers two surface types:
Spherical To calculate geometry over an Earth-like sphere, store your location data on a spherical surface and use
2dsphere (page 328) index.
Store your location data as GeoJSON objects with this coordinate-axis order: longitude, latitude. The coordinate
reference system for GeoJSON uses the WGS84 datum.
Flat To calculate distances on a Euclidean plane, store your location data as legacy coordinate pairs and use a 2d
(page 329) index.
Location Data If you choose spherical surface calculations, you store location data as either:
GeoJSON Objects Queries on GeoJSON objects always calculate on a sphere. The default coordinate reference
system for GeoJSON uses the WGS84 datum.

326

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

New in version 2.4: Support for GeoJSON storage and queries is new in version 2.4. Prior to version 2.4, all geospatial
data used coordinate pairs.
MongoDB supports the following GeoJSON objects:
Point
LineString
Polygon
Legacy Coordinate Pairs MongoDB supports spherical surface calculations on legacy coordinate pairs by converting the data to the GeoJSON Point type.
If you choose flat surface calculations, you can store data only as legacy coordinate pairs.
Query Operations MongoDBs geospatial query operators let you query for:
Inclusion MongoDB can query for locations contained entirely within a specified polygon. Inclusion queries use
the $geoWithin operator.
Intersection MongoDB can query for locations that intersect with a specified geometry. These queries apply only
to data on a spherical surface. These queries use the $geoIntersects operator.
Proximity MongoDB can query for the points nearest to another point. Proximity queries use the $near operator.
The $near operator requires a 2d or 2dsphere index.
Geospatial Indexes MongoDB provides the following geospatial index types to support the geospatial queries.
2dsphere 2dsphere (page 328) indexes support:
Calculations on a sphere
Both GeoJSON objects and legacy coordinate pairs
A compound index with scalar index fields (i.e. ascending or descending) as a prefix or suffix of the 2dsphere
index field
New in version 2.4: 2dsphere indexes are not available before version 2.4.
See also:
Query a 2dsphere Index (page 350)
2d 2d (page 329) indexes support:
Calculations using flat geometry
Legacy coordinate pairs (i.e., geospatial points on a flat coordinate system)
A compound index with only one additional field, as a suffix of the 2d index field
See also:
Query a 2d Index (page 353)

7.2. Index Concepts

327

MongoDB Documentation, Release 2.4.9

Geospatial Indexes and Sharding You cannot use a geospatial index as the shard key index.
You can create and maintain a geospatial index on a sharded collection if using different fields as the shard key.
Queries using $near are not supported for sharded collections. Use geoNear instead. You also can query for
geospatial data using $geoWithin.
Additional Resources The following pages provide complete documentation for geospatial indexes and queries:
2dsphere Indexes (page 328) A 2dsphere index supports queries that calculate geometries on an earth-like sphere.
The index supports data stored as both GeoJSON objects and as legacy coordinate pairs.
2d Indexes (page 329) The 2d index supports data stored as legacy coordinate pairs and is intended for use in MongoDB 2.2 and earlier.
Haystack Indexes (page 330) A haystack index is a special index optimized to return results over small areas. For
queries that use spherical geometry, a 2dsphere index is a better option than a haystack index.
2d Index Internals (page 331) Provides a more in-depth explanation of the internals of geospatial indexes. This material is not necessary for normal operations but may be useful for troubleshooting and for further understanding.
2dsphere Indexes New in version 2.4.
A 2dsphere index supports queries that calculate geometries on an earth-like sphere. The index supports data stored
as both GeoJSON objects and as legacy coordinate pairs. The index supports legacy coordinate pairs by converting
the data to the GeoJSON Point type.
The 2dsphere index supports all MongoDB geospatial queries: queries for inclusion, intersection and proximity.
A compound (page 322) 2dsphere index can reference multiple location and non-location fields within a collections
documents. You can arrange the fields in any order.
The default datum for an earth-like sphere in MongoDB 2.4 is WGS84. Coordinate-axis order is longitude, latitude.
See the http://docs.mongodb.org/manualreference/operator/query-geospatial for the
query operators that support geospatial queries.
Considerations MongoDB allows only one geospatial index per collection. You can create either a 2dsphere or
a 2d (page 329) per collection.
You cannot use a 2dsphere index as a shard key when sharding a collection. However, you can create and maintain
a geospatial index on a sharded collection by using a different field as the shard key.
GeoJSON Objects New in version 2.4.
MongoDB supports the following GeoJSON objects:
Point
LineString
Polygon
In order to index GeoJSON data, you must store the data in a location field that you name. The location field contains
a subdocument with a type field specifying the GeoJSON object type and a coordinates field specifying the
objects coordinates. Always store coordinates longitude, latitude order.
Use the following syntax:

328

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

{ <location field> : { type : "<GeoJSON type>" ,

coordinates : <coordinates>
} }

The following example stores a GeoJSON Point:

{ loc : { type : "Point" ,
coordinates : [ 40, 5 ]
} }

The following example stores a GeoJSON LineString:

{ loc : { type : "LineString" ,
coordinates : [ [ 40 , 5 ] , [ 41 , 6 ] ]
} }

Polygons consist of an array of GeoJSON LinearRing coordinate arrays. These LinearRings are closed
LineStrings. Closed LineStrings have at least four coordinate pairs and specify the same position as the
first and last coordinates.
The following example stores a GeoJSON Polygon with an exterior ring and no interior rings (or holes). Note the
first and last coordinate pair with the [ 0 , 0 ] coordinate:
{ loc :
{ type : "Polygon" ,
coordinates : [ [ [ 0 , 0 ] , [ 3 , 6 ] , [ 6 , 1 ] , [ 0 , 0 ] ] ]
} }

For Polygons with multiple rings:

The first described ring must be the exterior ring.
The exterior ring cannot self-intersect.
Any interior ring must be entirely contained by the outer ring.
Interior rings cannot intersect or overlap each other. Interior rings can share an edge.
The following document represents a polygon with an interior ring as GeoJSON:
{ loc :
{ type : "Polygon" ,
coordinates : [ [ [ 0 , 0 ] , [ 3 , 6 ] , [ 6 , 1 ] , [ 0 , 0 ] ],
[ [ 2 , 2 ] , [ 3 , 3 ] , [ 4 , 2 ] , [ 2 , 2 ] ] ]
} }

2d Indexes Use a 2d index for data stored as points on a two-dimensional plane. The 2d index is intended for
legacy coordinate pairs used in MongoDB 2.2 and earlier.
Use a 2d index if:
your database has legacy location data from MongoDB 2.2 or earlier, and
you do not intend to store any location data as GeoJSON objects.
See the http://docs.mongodb.org/manualreference/operator/query-geospatial for the
query operators that support geospatial queries.
Considerations MongoDB allows only one geospatial index per collection. You can create either a 2d or a 2dsphere
(page 328) per collection.

7.2. Index Concepts

329

MongoDB Documentation, Release 2.4.9

Figure 7.10: Diagram of a Polygon with internal ring.

Do not use a 2d index if your location data includes GeoJSON objects. To index on both legacy coordinate pairs and
GeoJSON objects, use a 2dsphere (page 328) index.
You cannot use a 2d index as a shard key when sharding a collection. However, you can create and maintain a
geospatial index on a sharded collection by using a different field as the shard key.
Behavior The 2d index supports calculations on a flat, Euclidean plane. The 2d index also supports distance-only
calculations on a sphere, but for geometric calculations (e.g. $geoWithin) on a sphere, store data as GeoJSON
objects and use the 2dsphere index type.
A 2d index can reference two fields. The first must be the location field. A 2d compound index constructs queries
that select first on the location field, and then filters those results by the additional criteria. A compound 2d index can
cover queries.
Points on a 2D Plane To store location data as legacy coordinate pairs, use an array or an embedded document.
When possible, use the array format:
loc : [ <longitude> , <latitude> ]

Consider the embedded document form:

loc : { lng : <longitude> , lat : <latitude> }

Arrays are preferred as certain languages do not guarantee associative map ordering.
For all points, if you use longitude and latitude, store coordinates in longitude, latitude order.
Haystack Indexes A haystack index is a special index that is optimized to return results over small areas. Haystack
indexes improve performance on queries that use flat geometry.

330

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

For queries that use spherical geometry, a 2dsphere index is a better option than a haystack index. 2dsphere indexes
(page 328) allow field reordering; haystack indexes require the first field to be the location field. Also, haystack indexes
are only usable via commands and so always return all results at once.
Haystack indexes create buckets of documents from the same geographic area in order to improve performance for
queries limited to that area. Each bucket in a haystack index contains all the documents within a specified proximity
to a given longitude and latitude.
To create a geohaystacks index, see Create a Haystack Index (page 355). For information and example on querying a
haystack index, see Query a Haystack Index (page 356).
2d Index Internals This document provides a more in-depth explanation of the internals of MongoDBs 2d geospatial indexes. This material is not necessary for normal operations or application development but may be useful for
troubleshooting and for further understanding.
Calculation of Geohash Values for 2d Indexes When you create a geospatial index on legacy coordinate pairs,
MongoDB computes geohash values for the coordinate pairs within the specified location range (page 353) and then
indexes the geohash values.
To calculate a geohash value, recursively divide a two-dimensional map into quadrants. Then assign each quadrant a
two-bit value. For example, a two-bit representation of four quadrants would be:
01

11

00

10

These two-bit values (00, 01, 10, and 11) represent each of the quadrants and all points within each quadrant. For
a geohash with two bits of resolution, all points in the bottom left quadrant would have a geohash of 00. The top
left quadrant would have the geohash of 01. The bottom right and top right would have a geohash of 10 and 11,
respectively.
To provide additional precision, continue dividing each quadrant into sub-quadrants. Each sub-quadrant would have
the geohash value of the containing quadrant concatenated with the value of the sub-quadrant. The geohash for the
upper-right quadrant is 11, and the geohash for the sub-quadrants would be (clockwise from the top left): 1101,
1111, 1110, and 1100, respectively.
Multi-location Documents for 2d Indexes New in version 2.0: Support for multiple locations in a document.
While 2d geospatial indexes do not support more than one set of coordinates in a document, you can use a multi-key
index (page 324) to index multiple coordinate pairs in a single document. In the simplest example you may have a
field (e.g. locs) that holds an array of coordinates, as in the following example:
{ _id : ObjectId(...),
locs : [ [ 55.5 , 42.3 ] ,
[ -74 , 44.74 ] ,
{ lng : 55.5 , lat : 42.3 } ]
}

The values of the array may be either arrays, as in [ 55.5, 42.3 ], or embedded documents, as in { lng :
55.5 , lat : 42.3 }.
You could then create a geospatial index on the locs field, as in the following:
db.places.ensureIndex( { "locs": "2d" } )

You may also model the location data as a field inside of a sub-document. In this case, the document would contain
a field (e.g. addresses) that holds an array of documents where each document has a field (e.g. loc:) that holds
location coordinates. For example:
7.2. Index Concepts

331

MongoDB Documentation, Release 2.4.9

{ _id : ObjectId(...),
name : "...",
addresses : [ {
context : "home" ,
loc : [ 55.5, 42.3 ]
} ,
{
context : "home",
loc : [ -74 , 44.74 ]
}
]
}

You could then create the geospatial index on the addresses.loc field as in the following example:
db.records.ensureIndex( { "addresses.loc": "2d" } )

For documents with multiple coordinate values, queries may return the same document multiple times if more than
one indexed coordinate pair satisfies the query constraints. Use the uniqueDocs parameter to geoNear or the
$uniqueDocs operator with $geoWithin.
To include the location field with the distance field in multi-location document queries, specify includeLocs:
true in the geoNear command.
See also:
geospatial-query-compatibility-chart
Text Indexes

New in version 2.4.

MongoDB provides text indexes to support text search of string content in documents of a collection. text indexes
are case-insensitive and can include any field whose value is a string or an array of string elements. You can only
access the text index with the text command.
Important:
Before you can create a text index or run the text command (page 333), you need to manually enable the text
search. See Enable Text Search (page 359) for information on how to enable the text search feature.
A collection can have at most one text index.

Create Text Index To create a text index, use the db.collection.ensureIndex() method. To index a
field that contains a string or an array of string elements, include the field and specify the string literal "text" in the
index document, as in the following example:
db.reviews.ensureIndex( { comments: "text" } )

For examples of creating text indexes on multiple fields, see Create a text Index (page 359).
text indexes drop language-specific stop words (e.g. in English, the, an, a, and, etc.) and uses simple
language-specific suffix stemming. See text-search-languages for the supported languages and Specify a Language for
Text Index (page 363) for details on specifying languages with text indexes.
text indexes can satisfy the filter component of a text search. For details, see Create text Index to Satisfy the
filter Component of Text Search (page 367).

332

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

Storage Requirements and Performance Costs text indexes have the following storage requirements and performance costs:
text indexes change the space allocation method for all future record allocations in a collection to
usePowerOf2Sizes.
text indexes can be large. They contain one index entry for each unique post-stemmed word in each indexed
field for each document inserted.
Building a text index is very similar to building a large multi-key index and will take longer than building a
simple ordered (scalar) index on the same data.
When building a large text index on an existing collection, ensure that you have a sufficiently high limit on
open file descriptors. See the recommended settings (page 223).
text indexes will impact insertion throughput because MongoDB must add an index entry for each unique
post-stemmed word in each indexed field of each new source document.
Additionally, text indexes do not store phrases or information about the proximity of words in the documents.
As a result, phrase queries will run much more effectively when the entire collection fits in RAM.
Text Search Text search supports the search of string content in documents of a collection. MongoDB provides the
text command to perform the text search. The text command accesses the text index.
The text search process:
tokenizes and stems the search term(s) during both the index creation and the text command execution.
assigns a score to each document that contains the search term in the indexed fields. The score determines the
relevance of a document to a given search query.
By default, the text command returns at most the top 100 matching documents as determined by the scores. The
command can search for words and phrases. The command matches on the complete stemmed words. For example, if
a document field contains the word blueberry, a search on the term blue will not match the document. However,
a search on either blueberry or blueberries will match.
For information and examples on various text search patterns, see Search String Content for Text (page 360).
Hashed Index

New in version 2.4.

Hashed indexes maintain entries with hashes of the values of the indexed field. The hashing function collapses subdocuments and computes the hash for the entire value but does not support multi-key (i.e. arrays) indexes.
Hashed indexes support sharding (page 479) a collection using a hashed shard key (page 493). Using a hashed shard
key to shard a collection ensures a more even distribution of data. See Shard a Collection Using a Hashed Shard Key
(page 513) for more details.
MongoDB can use the hashed index to support equality queries, but hashed indexes do not support range queries.
You may not create compound indexes that have hashed index fields or specify a unique constraint
on a hashed index; however, you can create both a hashed index and an ascending/descending
(i.e. non-hashed) index on the same field: MongoDB will use the scalar index for range queries.
Warning: MongoDB hashed indexes truncate floating point numbers to 64-bit integers before hashing. For
example, a hashed index would store the same value for a field that held a value of 2.3, 2.2, and 2.9. To
prevent collisions, do not use a hashed index for floating point numbers that cannot be reliably converted to
64-bit integers (and then back to floating point). MongoDB hashed indexes do not support floating point values
larger than 253 .

7.2. Index Concepts

333

MongoDB Documentation, Release 2.4.9

Create a hashed index using an operation that resembles the following:

db.active.ensureIndex( { a: "hashed" } )

This operation creates a hashed index for the active collection on the a field.

7.2.2 Index Properties

In addition to the numerous index types (page 319) MongoDB supports, indexes can also have various properties. The
following documents detail the index properties that can you can select when building an index.
TTL Indexes (page 334) The TTL index is used for TTL collections, which expire data after a period of time.
Unique Indexes (page 334) A unique index causes MongoDB to reject all documents that contain a duplicate value
for the indexed field.
Sparse Indexes (page 335) A sparse index does not index documents that do not have the indexed field.
TTL Indexes
TTL indexes are special indexes that MongoDB can use to automatically remove documents from a collection after
a certain amount of time. This is ideal for some types of information like machine generated event data, logs, and
session information that only need to persist in a database for a limited amount of time.
Considerations

TTL indexes have the following limitations:

Compound indexes (page 322) are not supported.
The indexed field must be a date type.
If the field holds an array, and there are multiple date-typed data in the index, the document will expire when
the lowest (i.e. earliest) matches the expiration threshold.
The TTL index does not guarantee that expired data will be deleted immediately. There may be a delay between the
time a document expires and the time that MongoDB removes the document from the database.
The background task that removes expired documents runs every 60 seconds. As a result, documents may remain in a
collection after they expire but before the background task runs or completes.
The duration of the removal operation depends on the workload of your mongod instance. Therefore, expired data
may exist for some time beyond the 60 second period between runs of the background task.
In all other respects, TTL indexes are normal indexes, and if appropriate, MongoDB can use these indexes to fulfill
arbitrary queries.
Additional Information

Expire Data from Collections by Setting TTL (page 163)

Unique Indexes
A unique index causes MongoDB to reject all documents that contain a duplicate value for the indexed field. To create
a unique index on the user_id field of the members collection, use the following operation in the mongo shell:

334

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

db.addresses.ensureIndex( { "user_id": 1 }, { unique: true } )

By default, unique is false on MongoDB indexes.

If you use the unique constraint on a compound index (page 322), then MongoDB will enforce uniqueness on the
combination of values rather than the individual value for any or all values of the key.
If a document does not have a value for the indexed field in a unique index, the index will store a null value for this
document. Because of the unique constraint, MongoDB will only permit one document that lacks the indexed field. If
there is more than one document without a value for the indexed field or is missing the indexed field, the index build
will fail with a duplicate key error.
You can combine the unique constraint with the sparse index (page 335) to filter these null values from the unique
index and avoid the error.
You may not specify a unique constraint on a hashed index (page 333).
Sparse Indexes
Sparse indexes only contain entries for documents that have the indexed field, even if the index field contains a null
value. The index skips over any document that is missing the indexed field. The index is sparse because it does not
include all documents of a collection. By contrast, non-sparse indexes contain all documents in a collection, storing
null values for those documents that do not contain the indexed field.
The following example in the mongo shell creates a sparse index on the xmpp_id field of the addresses collection:
db.addresses.ensureIndex( { "xmpp_id": 1 }, { sparse: true } )

By default, sparse is false on MongoDB indexes.

Warning: Using these indexes will sometimes result in incomplete results when filtering or sorting results,
because sparse indexes are not complete for all documents in a collection.
Note: Do not confuse sparse indexes in MongoDB with block-level3 indexes in other databases. Think of them as
dense indexes with a specific filter.
Tip
You can specify a sparse and unique index (page 334), that rejects documents that have duplicate values for a field,
but allows multiple documents that omit that key.

Examples

Sparse Index On A Collection Can Result In Incomplete Results Consider a collection scores that contains
the following documents:
{ "_id" : ObjectId("523b6e32fb408eea0eec2647"), "userid" : "newbie" }
{ "_id" : ObjectId("523b6e61fb408eea0eec2648"), "userid" : "abby", "score" : 82 }
{ "_id" : ObjectId("523b6e6ffb408eea0eec2649"), "userid" : "nina", "score" : 90 }

The collection has a sparse index on the field score:

3 http://en.wikipedia.org/wiki/Database_index#Sparse_index

7.2. Index Concepts

335

MongoDB Documentation, Release 2.4.9

db.scores.ensureIndex( { score: 1 } , { sparse: true } )

Then, the following query to return all documents in the scores collection sorted by the score field gives incomplete results:
db.scores.find().sort( { score: -1 } )

Because the document for the userid "newbie" does not contain the score field, the query, which uses the sparse
index, will return incomplete results that omit that document:
{ "_id" : ObjectId("523b6e6ffb408eea0eec2649"), "userid" : "nina", "score" : 90 }
{ "_id" : ObjectId("523b6e61fb408eea0eec2648"), "userid" : "abby", "score" : 82 }

Sparse Index with Unique Constraint Consider a collection scores that contains the following documents:
{ "_id" : ObjectId("523b6e32fb408eea0eec2647"), "userid" : "newbie" }
{ "_id" : ObjectId("523b6e61fb408eea0eec2648"), "userid" : "abby", "score" : 82 }
{ "_id" : ObjectId("523b6e6ffb408eea0eec2649"), "userid" : "nina", "score" : 90 }

You could create an index with a unique constraint (page 334) and sparse filter on the score field using the following
operation:
db.scores.ensureIndex( { score: 1 } , { sparse: true, unique: true } )

This index would permit inserting documents that had unique values for the score field or did not include a score
field. Consider the following insert operation (page 59):
db.scores.insert(
db.scores.insert(
db.scores.insert(
db.scores.insert(

{
{
{
{

"userid":
"userid":
"userid":
"userid":

"PWWfO8lFs1", "score": 43 } )
"XlSOX66gEy", "score": 34 } )
"nuZHu2tcRm" } )
"HIGvEZfdc5" } )

However, this index would not permit adding the following documents:
db.scores.insert( { "userid": "PWWfO8lFs1", "score": 82 } )
db.scores.insert( { "userid": "XlSOX66gEy", "score": 90 } )

7.2.3 Index Creation

MongoDB provides several options that only affect the creation of the index. Specify these options in a document as
the second argument to the db.collection.ensureIndex() method. This section describes the uses of these
creation options and their behavior.
Related
Some options that you can specify to ensureIndex() options control the properties of the index (page 334), which
are not index creation options. For example, the unique (page 334) option affects the behavior of the index after
creation.
For a detailed description of MongoDBs index types, see Index Types (page 319) and Index Properties (page 334) for
related documentation.

Background Construction
By default, creating an index blocks all other operations on a database. When building an index on a collection, the
database that holds the collection is unavailable for read or write operations until the index build completes. Any
336

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

operation that requires a read or write lock on all databases (e.g. listDatabases) will wait for the foreground index
build to complete.
For potentially long running index building operations, consider the background operation so that the MongoDB
database remains available during the index building operation. For example, to create an index in the background of
the zipcode field of the people collection, issue the following:
db.people.ensureIndex( { zipcode: 1}, {background: true} )

By default, background is false for building MongoDB indexes.

You can combine the background option with other options, as in the following:
db.people.ensureIndex( { zipcode: 1}, {background: true, sparse: true } )

Behavior

As of MongoDB version 2.4, a mongod instance can build more than one index in the background concurrently.
Changed in version 2.4: Before 2.4, a mongod instance could only build one background index per database at a time.
Changed in version 2.2: Before 2.2, a single mongod instance could only build one index at a time.
Background indexing operations run in the background so that other database operations can run while creating the
index. However, the mongo shell session or connection where you are creating the index will block until the index
build is complete. To continue issuing commands to the database, open another connection or mongo instance.
Queries will not use partially-built indexes: the index will only be usable once the index build is complete.
Note:
If MongoDB is building an index in the background, you cannot perform other administrative operations involving that collection, including running repairDatabase, dropping the collection (i.e.
db.collection.drop()), and running compact. These operations will return an error during background
index builds.

Performance

The background index operation uses an incremental approach that is slower than the normal foreground index
builds. If the index is larger than the available RAM, then the incremental process can take much longer than the
foreground build.
If your application includes ensureIndex() operations, and an index doesnt exist for other operational concerns,
building the index can have a severe impact on the performance of the database.
To avoid performance issues, make sure that your application checks for the indexes at start up using the
getIndexes() method or the equivalent method for your driver4 and terminates if the proper indexes do not exist. Always build indexes in production instances using separate application code, during designated maintenance
windows.
Building Indexes on Secondaries

Background index operations on a replica set primary become foreground indexing operations on secondary members
of the set. All indexing operations on secondaries block replication.
4 http://api.mongodb.org/

7.2. Index Concepts

337

MongoDB Documentation, Release 2.4.9

To build large indexes on secondaries the best approach is to restart one secondary at a time in standalone mode and
build the index. After building the index, restart as a member of the replica set, allow it to catch up with the other
members of the set, and then build the index on the next secondary. When all the secondaries have the new index, step
down the primary, restart it as a standalone, and build the index on the former primary.
Remember, the amount of time required to build the index on a secondary must be within the window of the oplog, so
that the secondary can catch up with the primary.
Indexes on secondary members in recovering mode are always built in the foreground to allow them to catch up as
soon as possible.
See Build Indexes on Replica Sets (page 344) for a complete procedure for building indexes on secondaries.
Drop Duplicates
MongoDB cannot create a unique index (page 334) on a field that has duplicate values. To force the creation of a
unique index, you can specify the dropDups option, which will only index the first occurrence of a value for the key,
and delete all subsequent values.
Important: As in all unique indexes, if a document does not have the indexed field, MongoDB will include it in the
index with a null value.
If subsequent fields do not have the indexed field, and you have set {dropDups: true}, MongoDB will remove
these documents from the collection when creating the index. If you combine dropDups with the sparse (page 335)
option, this index will only include documents in the index that have the value, and the documents without the field
will remain in the database.
To create a unique index that drops duplicates on the username field of the accounts collection, use a command
in the following form:
db.accounts.ensureIndex( { username: 1 }, { unique: true, dropDups: true } )

Warning: Specifying { dropDups:

tion.

true } will delete data from your database. Use with extreme cau-

By default, dropDups is false.

Index Names
The default name for an index is the concatenation of the indexed keys and each keys direction in the index, 1 or -1.
Example
Issue the following command to create an index on item and quantity:
db.products.ensureIndex( { item: 1, quantity: -1 } )

The resulting index is named: item_1_quantity_-1.

Optionally, you can specify a name for an index instead of using the default name.
Example
Issue the following command to create an index on item and quantity and specify inventory as the index
name:

338

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

db.products.ensureIndex( { item: 1, quantity: -1 } , { name: "inventory" } )

The resulting index has the name inventory.

To view the name of an index, use the getIndexes() method.

7.3 Indexing Tutorials

Indexes allow MongoDB to process and fulfill queries quickly by creating small and efficient representations of the
documents in a collection.
The documents in this section outline specific tasks related to building and maintaining indexes for data in MongoDB
collections and discusses strategies and practical approaches. For a conceptual overview of MongoDB indexing, see
the Index Concepts (page 318) document.
Index Creation Tutorials (page 339) Create and configure different types of indexes for different purposes.
Index Management Tutorials (page 346) Monitor and assess index performance and rebuild indexes as needed.
Geospatial Index Tutorials (page 350) Create indexes that support data stored as GeoJSON objects and legacy coordinate pairs.
Text Search Tutorials (page 358) Build and configure indexes that support full-text searches.
Indexing Strategies (page 368) The factors that affect index performance and practical approaches to indexing in
MongoDB

7.3.1 Index Creation Tutorials

Instructions for creating and configuring indexes in MongoDB and building indexes on replica sets and sharded clusters.
Create an Index (page 340) Build an index for any field on a collection.
Create a Compound Index (page 341) Build an index of multiple fields on a collection.
Create a Unique Index (page 341) Build an index that enforces unique values for the indexed field or fields.
Create a Sparse Index (page 342) Build an index that omits references to documents that do not include the indexed
field. This saves space when indexing fields that are present in only some documents.
Create a Hashed Index (page 343) Compute a hash of the value of a field in a collection and index the hashed value.
These indexes permit equality queries and may be suitable shard keys for some collections.
Build Indexes on Replica Sets (page 344) To build indexes on a replica set, you build the indexes separately on the
primary and the secondaries, as described here.
Build Indexes in the Background (page 345) Background index construction allows read and write operations to
continue while building the index, but take longer to complete and result in a larger index.
Build Old Style Indexes (page 346) A {v :
2.0 (or later) to MongoDB version 1.8.

7.3. Indexing Tutorials

0} index is necessary if you need to roll back from MongoDB version

339

MongoDB Documentation, Release 2.4.9

Create an Index
Indexes allow MongoDB to process and fulfill queries quickly by creating small and efficient representations of the
documents in a collection. MongoDB creates an index on the _id field of every collection by default, but allows users
to create indexes for any collection using on any field in a document.
This tutorial describes how to create an index on a single field. MongoDB also supports compound indexes (page 322),
which are indexes on multiple fields. See Create a Compound Index (page 341) for instructions on building compound
indexes.
Create an Index on a Single Field

To create an index, use ensureIndex() or a similar method from your driver5 . For example the following creates
an index on the phone-number field of the people collection:
db.people.ensureIndex( { "phone-number": 1 } )

ensureIndex() only creates an index if an index of the same specification does not already exist.
All indexes support and optimize the performance for queries that select on this field. For queries that cannot use an
index, MongoDB must scan all documents in a collection for documents that match the query.
Tip
The value of the field in the index specification describes the kind of index for that field. For example, a value of 1
specifies an index that orders items in ascending order. A value of -1 specifies an index that orders items in descending
order.

Examples

If you create an index on the user_id field in the records, this index is, the index will support the following
query:
db.records.find( { user_id: 2 } )

However, the following query, on the profile_url field is not supported by this index:
db.records.find( { profile_url: 2 } )

Additional Considerations

If your collection holds a large amount of data, and your application needs to be able to access the data while building
the index, consider building the index in the background, as described in Background Construction (page 336). To
build indexes on replica sets, see the Build Indexes on Replica Sets (page 344) section for more information.
Note: To build or rebuild indexes for a replica set see Build Indexes on Replica Sets (page 344).
Some drivers may specify indexes, using NumberLong(1) rather than 1 as the specification. This does not have any
affect on the resulting index.
See also:
5 http://api.mongodb.org/

340

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

Create a Compound Index (page 341), Indexing Tutorials (page 339) and Index Concepts (page 318) for more information.
Create a Compound Index
Indexes allow MongoDB to process and fulfill queries quickly by creating small and efficient representations of the
documents in a collection. MongoDB supports indexes that include content on a single field, as well as compound
indexes (page 322) that include content from multiple fields. Continue reading for instructions and examples of
building a compound index.
Build a Compound Index

To create a compound index (page 322) use an operation that resembles the following prototype:
db.collection.ensureIndex( { a: 1, b: 1, c: 1 } )

Example

The following operation will create an index on the item, category, and price fields of the products collection:
db.products.ensureIndex( { item: 1, category: 1, price: 1 } )

Additional Considerations

If your collection holds a large amount of data, and your application needs to be able to access the data while building
the index, consider building the index in the background, as described in Background Construction (page 336). To
build indexes on replica sets, see the Build Indexes on Replica Sets (page 344) section for more information.
Note: To build or rebuild indexes for a replica set see Build Indexes on Replica Sets (page 344).
Some drivers may specify indexes, using NumberLong(1) rather than 1 as the specification. This does not have any
affect on the resulting index.
Tip
The value of the field in the index specification describes the kind of index for that field. For example, a value of 1
specifies an index that orders items in ascending order. A value of -1 specifies an index that orders items in descending
order.
See also:
Create an Index (page 340), Indexing Tutorials (page 339) and Index Concepts (page 318) for more information.
Create a Unique Index
MongoDB allows you to specify a unique constraint (page 334) on an index. These constraints prevent applications
from inserting documents that have duplicate values for the inserted fields. Additionally, if you want to create an index
on a collection that has existing data that might have duplicate values for the indexed field, you may choose to combine
unique enforcement with duplicate dropping (page 338).

7.3. Indexing Tutorials

341

MongoDB Documentation, Release 2.4.9

Unique Indexes

To create a unique indexes (page 334), consider the following prototype:

db.collection.ensureIndex( { a: 1 }, { unique: true } )

For example, you may want to create a unique index on the "tax-id": of the accounts collection to prevent
storing multiple account records for the same legal entity:
db.accounts.ensureIndex( { "tax-id": 1 }, { unique: true } )

The _id index (page 321) is a unique index. In some situations you may consider using _id field itself for this kind of
data rather than using a unique index on another field.
In many situations you will want to combine the unique constraint with the sparse option. When MongoDB
indexes a field, if a document does not have a value for a field, the index entry for that item will be null. Since
unique indexes cannot have duplicate values for a field, without the sparse option, MongoDB will reject the second
document and all subsequent documents without the indexed field. Consider the following prototype.
db.collection.ensureIndex( { a: 1 }, { unique: true, sparse: true } )

You can also enforce a unique constraint on compound indexes (page 322), as in the following prototype:
db.collection.ensureIndex( { a: 1, b: 1 }, { unique: true } )

These indexes enforce uniqueness for the combination of index keys and not for either key individually.
Drop Duplicates

To force the creation of a unique index (page 334) index on a collection with duplicate values in the field you are
indexing you can use the dropDups option. This will force MongoDB to create a unique index by deleting documents
with duplicate values when building the index. Consider the following prototype invocation of ensureIndex():
db.collection.ensureIndex( { a: 1 }, { unique: true, dropDups: true } )

See the full documentation of duplicate dropping (page 338) for more information.
Warning: Specifying { dropDups:
tion.

true } may delete data from your database. Use with extreme cau-

Refer to the ensureIndex() documentation for additional index creation options.

Create a Sparse Index
Sparse indexes are like non-sparse indexes, except that they omit references to documents that do not include the
indexed field. For fields that are only present in some documents sparse indexes may provide a significant space
savings. See Sparse Indexes (page 335) for more information about sparse indexes and their use.
See also:
Index Concepts (page 318) and Indexing Tutorials (page 339) for more information.
Prototype

To create a sparse index (page 335) on a field, use an operation that resembles the following prototype:

342

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

db.collection.ensureIndex( { a: 1 }, { sparse: true } )

Example

The following operation, creates a sparse index on the users collection that only includes a document in the index if
the twitter_name field exists in a document.
db.users.ensureIndex( { twitter_name: 1 }, { sparse: true } )

The index excludes all documents that do not include the twitter_name field.
Considerations

Note: Sparse indexes can affect the results returned by the query, particularly with respect to sorts on fields not
included in the index. See the sparse index (page 335) section for more information.

Create a Hashed Index

New in version 2.4.
Hashed indexes (page 333) compute a hash of the value of a field in a collection and index the hashed value. These
indexes permit equality queries and may be suitable shard keys for some collections.
Tip
MongoDB automatically computes the hashes when resolving queries using hashed indexes. Applications do not need
to compute hashes.
See
Hashed Shard Keys (page 493) for more information about hashed indexes in sharded clusters, as well as Index Concepts (page 318) and Indexing Tutorials (page 339) for more information about indexes.

Procedure

To create a hashed index (page 333), specify hashed as the value of the index key, as in the following example:
Example
Specify a hashed index on _id
db.collection.ensureIndex( { _id: "hashed" } )

Considerations

MongoDB supports hashed indexes of any single field. The hashing function collapses sub-documents and computes
the hash for the entire value, but does not support multi-key (i.e. arrays) indexes.
You may not create compound indexes that have hashed index fields.

7.3. Indexing Tutorials

343

MongoDB Documentation, Release 2.4.9

Build Indexes on Replica Sets

Background index creation operations (page 336) become foreground indexing operations on secondary members of
replica sets. The foreground index building process blocks all replication and read operations on the secondaries while
they build the index.
Secondaries will begin building indexes after the primary finishes building the index. In sharded clusters, the mongos
will send ensureIndex() to the primary members of the replica set for each shard, which then replicate to the
secondaries after the primary finishes building the index.
To minimize the impact of building an index on your replica set, use the following procedure to build indexes on
secondaries:
See
Indexing Tutorials (page 339) and Index Concepts (page 318) for more information.

Considerations

Ensure that your oplog is large enough to permit the indexing or re-indexing operation to complete without
falling too far behind to catch up. See the oplog sizing (page 411) documentation for additional information.
This procedure does take one member out of the replica set at a time. However, this procedure will only affect
one member of the set at a time rather than all secondaries at the same time.
Do not use this procedure when building a unique index (page 334) with the dropDups option.
Procedure

Note: If you need to build an index in a sharded cluster, repeat the following procedure for each replica set that
provides each shard.

Stop One Secondary Stop the mongod process on one secondary. Restart the mongod process without the
--replSet option and running on a different port. 6 This instance is now in standalone mode.
For example, if your mongod normally runs with on the default port of 27017 with the --replSet option you
would use the following invocation:
mongod --port 47017

Build the Index Create the new index using the ensureIndex() in the mongo shell, or comparable method in
your driver. This operation will create or rebuild the index on this mongod instance
For example, to create an ascending index on the username field of the records collection, use the following
mongo shell operation:
db.records.ensureIndex( { username: 1 } )

See also:
Create an Index (page 340) and Create a Compound Index (page 341) for more information.
6 By running the mongod on a different port, you ensure that the other members of the replica set and all clients will not contact the member
while you are building the index.

344

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

Restart the Program mongod When the index build completes, start the mongod instance with the --replSet
option on its usual port:
mongod --port 27017 --replSet rs0

Modify the port number (e.g. 27017) or the replica set name (e.g. rs0) as needed.
Allow replication to catch up on this member.
Build Indexes on all Secondaries For each secondary in the set, build an index according to the following steps:
1. Stop One Secondary (page 344)
2. Build the Index (page 344)
3. Restart the Program mongod (page 345)
Build the Index on the Primary To build an index on the primary you can either:
1. Build the index in the background (page 345) on the primary.
2. Step down the primary using the rs.stepDown() method in the mongo shell to cause the current primary to
become a secondary graceful and allow the set to elect another member as primary.
Then repeat the index building procedure, listed below, to build the index on the primary:
(a) Stop One Secondary (page 344)
(b) Build the Index (page 344)
(c) Restart the Program mongod (page 345)
Building the index on the background, takes longer than the foreground index build and results in a less compact index
structure. Additionally, the background index build may impact write performance on the primary. However, building
the index in the background allows the set to be continuously up for write operations during while MongoDB builds
the index.
Build Indexes in the Background
By default, MongoDB builds indexes in the foreground and prevent all read and write operations to the database while
the index builds. Also, no operation that requires a read or write lock on all databases (e.g. listDatabases) can occur
during a foreground index build.
Background index construction (page 336) allows read and write operations to continue while building the index.
See also:
Index Concepts (page 318) and Indexing Tutorials (page 339) for more information.
Considerations

Background index builds take longer to complete and result in an index that is initially larger, or less compact, than an
index built in the foreground. Overtime the compactness of indexes built in the background will approach foregroundbuilt indexes.
After MongoDB finishes building the index, background-built indexes are functionally identical to any other index.

7.3. Indexing Tutorials

345

MongoDB Documentation, Release 2.4.9

Procedure

To create an index in the background, add the background argument to the ensureIndex() operation, as in the
following index:
db.collection.ensureIndex( { a: 1 }, { background: true } )

Consider the section on background index construction (page 336) for more information about these indexes and their
implications.
Build Old Style Indexes
Important: Use this procedure only if you must have indexes that are compatible with a version of MongoDB earlier
than 2.0.
MongoDB version 2.0 introduced the {v:1} index format. MongoDB versions 2.0 and later support both the {v:1}
format and the earlier {v:0} format.
MongoDB versions prior to 2.0, however, support only the {v:0} format. If you need to roll back MongoDB to a
version prior to 2.0, you must drop and re-create your indexes.
To build pre-2.0 indexes, use the dropIndexes() and ensureIndex() methods. You cannot simply reindex the
collection. When you reindex on versions that only support {v:0} indexes, the v fields in the index definition still
hold values of 1, even though the indexes would now use the {v:0} format. If you were to upgrade again to version
2.0 or later, these indexes would not work.
Example
Suppose you rolled back from MongoDB 2.0 to MongoDB 1.8, and suppose you had the following index on the
items collection:
{ "v" : 1, "key" : { "name" : 1 }, "ns" : "mydb.items", "name" : "name_1" }

The v field tells you the index is a {v:1} index, which is incompatible with version 1.8.
To drop the index, issue the following command:
db.items.dropIndex( { name : 1 } )

To recreate the index as a {v:0} index, issue the following command:

db.foo.ensureIndex( { name : 1 } , { v : 0 } )

See also:
Index Performance Enhancements (page 620).

7.3.2 Index Management Tutorials

Instructions for managing indexes and assessing index performance and use.
Remove Indexes (page 347) Drop an index from a collection.
Rebuild Indexes (page 347) In a single operation, drop all indexes on a collection and then rebuild them.
Manage In-Progress Index Creation (page 348) Check the status of indexing progress, or terminate an ongoing index build.

346

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

Return a List of All Indexes (page 348) Obtain a list of all indexes on a collection or of all indexes on all collections
in a database.
Measure Index Use (page 349) Study query operations and observe index use for your database.
Remove Indexes
To remove an index from a collection use the dropIndex() method and the following procedure. If you simply
need to rebuild indexes you can use the process described in the Rebuild Indexes (page 347) document.
See also:
Indexing Tutorials (page 339) and Index Concepts (page 318) for more information about indexes and indexing operations in MongoDB.
Operations

To remove an index, use the db.collection.dropIndex() method, as in the following example:

db.accounts.dropIndex( { "tax-id": 1 } )

This will remove the index on the "tax-id" field in the accounts collection. The shell provides the following
document after completing the operation:
{ "nIndexesWas" : 3, "ok" : 1 }

Where the value of nIndexesWas reflects the number of indexes before removing this index. You can also use the
db.collection.dropIndexes() to remove all indexes, except for the _id index (page 321) from a collection.
These shell helpers provide wrappers around the dropIndexes database command. Your client library (page 95)
may have a different or additional interface for these operations.
Rebuild Indexes
If you need to rebuild indexes for a collection you can use the db.collection.reIndex() method to rebuild all
indexes on a collection in a single operation. This operation drops all indexes, including the _id index (page 321), and
then rebuilds all indexes.
See also:
Index Concepts (page 318) and Indexing Tutorials (page 339).
Process

The operation takes the following form:

db.accounts.reIndex()

MongoDB will return the following document when the operation completes:
{
"nIndexesWas" : 2,
"msg" : "indexes dropped for collection",
"nIndexes" : 2,
"indexes" : [
{

7.3. Indexing Tutorials

347

MongoDB Documentation, Release 2.4.9

"key" : {
"_id" : 1,
"tax-id" : 1
},
"ns" : "records.accounts",
"name" : "_id_"
}
],
"ok" : 1
}

This shell helper provides a wrapper around the reIndex database command. Your client library (page 95) may
have a different or additional interface for this operation.
Additional Considerations

Note: To build or rebuild indexes for a replica set see Build Indexes on Replica Sets (page 344).

Manage In-Progress Index Creation

To see the status of the indexing processes, you can use the db.currentOp() method in the mongo shell. The
value of the query field and the msg field will indicate if the operation is an index build. The msg field also indicates
the percent of the build that is complete.
To terminate an ongoing index build, use the db.killOp() method in the mongo shell.
For more information see db.currentOp().
Changed in version 2.4: Before MongoDB 2.4, you could only terminate background index builds. After 2.4, you can
terminate any index build, including foreground index builds.
Return a List of All Indexes
When performing maintenance you may want to check which indexes exist on a collection. Every index on a collection
has a corresponding document in the system.indexes (page 227) collection, and you can use standard queries (i.e.
find()) to list the indexes, or in the mongo shell, the getIndexes() method to return a list of the indexes on a
collection, as in the following examples.
See also:
Index Concepts (page 318) and Indexing Tutorials (page 339) for more information about indexes in MongoDB and
common index management operations.
List all Indexes on a Collection

To return a list of all indexes on a collection, use the db.collection.getIndexes() method or a similar
method for your driver7 .
For example, to view all indexes on the people collection:
db.people.getIndexes()
7 http://api.mongodb.org/

348

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

List all Indexes for a Database

To return a list of all indexes on all collections in a database, use the following operation in the mongo shell:
db.system.indexes.find()

See system.indexes (page 227) for more information about these documents.
Measure Index Use
Synopsis

Query performance is a good general indicator of index use; however, for more precise insight into index use, MongoDB provides a number of tools that allow you to study query operations and observe index use for your database.
See also:
Index Concepts (page 318) and Indexing Tutorials (page 339) for more information.
Operations

Return Query Plan with explain() Append the explain() method to any cursor (e.g. query) to return a
document with statistics about the query process, including the index used, the number of documents scanned, and the
time the query takes to process in milliseconds.
Control Index Use with hint() Append the hint() to any cursor (e.g. query) with the index as the argument to
force MongoDB to use a specific index to fulfill the query. Consider the following example:
db.people.find( { name: "John Doe", zipcode: { $gt: 63000 } } } ).hint( { zipcode: 1 } )

You can use hint() and explain() in conjunction with each other to compare the effectiveness of a specific
index. Specify the $natural operator to the hint() method to prevent MongoDB from using any index:
db.people.find( { name: "John Doe", zipcode: { $gt: 63000 } } } ).hint( { $natural: 1 } )

Instance Index Use Reporting MongoDB provides a number of metrics of index use and operation that you may
want to consider when analyzing index use for your database:
In the output of serverStatus:
indexCounters
scanned
scanAndOrder
In the output of collStats:
totalIndexSize
indexSizes
In the output of dbStats:
dbStats.indexes
dbStats.indexSize

7.3. Indexing Tutorials

349

MongoDB Documentation, Release 2.4.9

7.3.3 Geospatial Index Tutorials

Instructions for creating and querying 2d, 2dsphere, and haystack indexes.
Create a 2dsphere Index (page 350) A 2dsphere index supports data stored as both GeoJSON objects and as
legacy coordinate pairs.
Query a 2dsphere Index (page 350) Search for locations within, near, or intersected by a GeoJSON shape, or within
a circle as defined by coordinate points on a sphere.
Create a 2d Index (page 352) Create a 2d index to support queries on data stored as legacy coordinate pairs.
Query a 2d Index (page 353) Search for locations using legacy coordinate pairs.
Create a Haystack Index (page 355) A haystack index is optimized to return results over small areas. For queries
that use spherical geometry, a 2dsphere index is a better option.
Query a Haystack Index (page 356) Search based on location and non-location data within a small area.
Calculate Distance Using Spherical Geometry (page 356) Convert distances to radians and back again.
Create a 2dsphere Index
To create a geospatial index for GeoJSON-formatted data, use the ensureIndex() method and set the value of the
location field for your collection to 2dsphere. A 2dsphere index can be a compound index (page 322) and does
not require the location field to be the first field indexed.
To create the index use the following syntax:
db.points.ensureIndex( { <location field> : "2dsphere" } )

The following are four example commands for creating a 2dsphere index:
db.points.ensureIndex(
db.points.ensureIndex(
db.points.ensureIndex(
db.points.ensureIndex(

{
{
{
{

loc : "2dsphere"
loc : "2dsphere"
rating : 1 , loc
loc : "2dsphere"

}
,
:
,

)
type : 1 } )
"2dsphere" } )
rating : 1 , category : -1 } )

The first example creates a simple geospatial index on the location field loc. The second example creates a compound
index where the second field contains non-location data. The third example creates an index where the location field
is not the primary field: the location field does not have to be the first field in a 2dsphere index. The fourth example
creates a compound index with three fields. You can include as many fields as you like in a 2dsphere index.
Query a 2dsphere Index
The following sections describe queries supported by the 2dsphere index. For an overview of recommended geospatial queries, see geospatial-query-compatibility-chart.
GeoJSON Objects Bounded by a Polygon

The $geoWithin operator queries for location data found within a GeoJSON polygon. Your location data must be
stored in GeoJSON format. Use the following syntax:
db.<collection>.find( { <location field> :
{ $geoWithin :
{ $geometry :
{ type : "Polygon" ,

350

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

coordinates : [ <coordinates> ]
} } } } )

The following example selects all points and shapes that exist entirely within a GeoJSON polygon:
db.places.find( { loc :
{ $geoWithin :
{ $geometry :
{ type : "Polygon" ,
coordinates : [ [
[ 0 , 0 ] ,
[ 3 , 6 ] ,
[ 6 , 1 ] ,
[ 0 , 0 ]
] ]
} } } } )

Intersections of GeoJSON Objects

New in version 2.4.

The $geoIntersects operator queries for locations that intersect a specified GeoJSON object. A location intersects the object if the intersection is non-empty. This includes documents that have a shared edge.
The $geoIntersects operator uses the following syntax:
db.<collection>.find( { <location field> :
{ $geoIntersects :
{ $geometry :
{ type : "<GeoJSON object type>" ,
coordinates : [ <coordinates> ]
} } } } )

The following example uses $geoIntersects to select all indexed points and shapes that intersect with the polygon
defined by the coordinates array.
db.places.find( { loc :
{ $geoIntersects :
{ $geometry :
{ type : "Polygon" ,
coordinates: [ [
[ 0 , 0 ] ,
[ 3 , 6 ] ,
[ 6 , 1 ] ,
[ 0 , 0 ]
] ]
} } } } )

Proximity to a GeoJSON Point

Proximity queries return the points closest to the defined point and sorts the results by distance. A proximity query on
GeoJSON data requires a 2dsphere index.
To query for proximity to a GeoJSON point, use either the $near operator or geoNear command. Distance is in
meters.
The $near uses the following syntax:

7.3. Indexing Tutorials

351

MongoDB Documentation, Release 2.4.9

db.<collection>.find( { <location field> :

{ $near :
{ $geometry :
{ type : "Point" ,
coordinates : [ <longitude> , <latitude> ] } ,
$maxDistance : <distance in meters>
} } } )

For examples, see $near.

The geoNear command uses the following syntax:
db.runCommand( { geoNear : <collection> ,
near : { type : "Point" ,
coordinates: [ <longitude>, <latitude> ] } ,
spherical : true } )

The geoNear command offers more options and returns more information than does the $near operator. To run the
command, see geoNear.
Points within a Circle Defined on a Sphere

To select all grid coordinates in a spherical cap on a sphere, use $geoWithin with the $centerSphere operator.
Specify an array that contains:
The grid coordinates of the circles center point
The circles radius measured in radians. To calculate radians, see Calculate Distance Using Spherical Geometry
(page 356).
Use the following syntax:
db.<collection>.find( { <location field> :
{ $geoWithin :
{ $centerSphere :
[ [ <x>, <y> ] , <radius> ] }
} } )

The following example queries grid coordinates and returns all documents within a 10 mile radius of longitude 88 W
and latitude 30 N. The example converts the distance, 10 miles, to radians by dividing by the approximate radius of
the earth, 3959 miles:
db.places.find( { loc :
{ $geoWithin :
{ $centerSphere :
[ [ 88 , 30 ] , 10 / 3959 ]
} } } )

Create a 2d Index
To build a geospatial 2d index, use the ensureIndex() method and specify 2d. Use the following syntax:
db.<collection>.ensureIndex( { <location field> : "2d" ,
<additional field> : <value> } ,
{ <index-specification options> } )

The 2d index uses the following optional index-specification options:

352

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

{ min : <lower bound> , max : <upper bound> ,

bits : <bit precision> }

Define Location Range for a 2d Index

By default, a 2d index assumes longitude and latitude and has boundaries of -180 inclusive and 180 non-inclusive
(i.e. [ -180 , 180 ]). If documents contain coordinate data outside of the specified range, MongoDB returns an
error.
Important: The default boundaries allow applications to insert documents with invalid latitudes greater than 90 or
less than -90. The behavior of geospatial queries with such invalid points is not defined.
On 2d indexes you can change the location range.
You can build a 2d geospatial index with a location range other than the default. Use the min and max options when
creating the index. Use the following syntax:
db.collection.ensureIndex( { <location field> : "2d" } ,
{ min : <lower bound> , max : <upper bound> } )

Define Location Precision for a 2d Index

By default, a 2d index on legacy coordinate pairs uses 26 bits of precision, which is roughly equivalent to 2 feet or 60
centimeters of precision using the default range of -180 to 180. Precision is measured by the size in bits of the geohash
values used to store location data. You can configure geospatial indexes with up to 32 bits of precision.
Index precision does not affect query accuracy. The actual grid coordinates are always used in the final query processing. Advantages to lower precision are a lower processing overhead for insert operations and use of less space. An
advantage to higher precision is that queries scan smaller portions of the index to return results.
To configure a location precision other than the default, use the bits option when creating the index. Use following
syntax:
db.<collection>.ensureIndex( {<location field> : "<index type>"} ,
{ bits : <bit precision> } )

For information on the internals of geohash values, see Calculation of Geohash Values for 2d Indexes (page 331).
Query a 2d Index
The following sections describe queries supported by the 2d index. For an overview of recommended geospatial
queries, see geospatial-query-compatibility-chart.
Points within a Shape Defined on a Flat Surface

To select all legacy coordinate pairs found within a given shape on a flat surface, use the $geoWithin operator along
with a shape operator. Use the following syntax:
db.<collection>.find( { <location field> :
{ $geoWithin :
{ $box|$polygon|$center : <coordinates>
} } } )

7.3. Indexing Tutorials

353

MongoDB Documentation, Release 2.4.9

The following queries for documents within a rectangle defined by [ 0 , 0 ] at the bottom left corner and by [
100 , 100 ] at the top right corner.
db.places.find( { loc :
{ $geoWithin :
{ $box : [ [ 0 , 0 ] ,
[ 100 , 100 ] ]
} } } )

The following queries for documents that are within the circle centered on [ -74 , 40.74 ] and with a radius of
10:
db.places.find( { loc: { $geoWithin :
{ $center : [ [-74, 40.74 ] , 10 ]
} } } )

For syntax and examples for each shape, see the following:
$box
$polygon
$center (defines a circle)
Points within a Circle Defined on a Sphere

MongoDB supports rudimentary spherical queries on flat 2d indexes for legacy reasons. In general, spherical calculations should use a 2dsphere index, as described in 2dsphere Indexes (page 328).
To query for legacy coordinate pairs in a spherical cap on a sphere, use $geoWithin with the $centerSphere
operator. Specify an array that contains:
The grid coordinates of the circles center point
The circles radius measured in radians. To calculate radians, see Calculate Distance Using Spherical Geometry
(page 356).
Use the following syntax:
db.<collection>.find( { <location field> :
{ $geoWithin :
{ $centerSphere : [ [ <x>, <y> ] , <radius> ] }
} } )

The following example query returns all documents within a 10-mile radius of longitude 88 W and latitude 30 N.
The example converts distance to radians by dividing distance by the approximate radius of the earth, 3959 miles:
db.<collection>.find( { loc : { $geoWithin :
{ $centerSphere :
[ [ 88 , 30 ] , 10 / 3959 ]
} } } )

Proximity to a Point on a Flat Surface

Proximity queries return the 100 legacy coordinate pairs closest to the defined point and sort the results by distance.
Use either the $near operator or geoNear command. Both require a 2d index.
The $near operator uses the following syntax:

354

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

db.<collection>.find( { <location field> :

{ $near : [ <x> , <y> ]
} } )

For examples, see $near.

The geoNear command uses the following syntax:
db.runCommand( { geoNear: <collection>, near: [ <x> , <y> ] } )

The geoNear command offers more options and returns more information than does the $near operator. To run the
command, see geoNear.
Exact Matches on a Flat Surface

You can use the db.collection.find() method to query for an exact match on a location. These queries use
the following syntax:
db.<collection>.find( { <location field>: [ <x> , <y> ] } )

This query will return any documents with the value of [ <x> , <y> ].
Create a Haystack Index
To build a haystack index, use the bucketSize option when creating the index. A bucketSize of 5 creates an
index that groups location values that are within 5 units of the specified longitude and latitude. The bucketSize also
determines the granularity of the index. You can tune the parameter to the distribution of your data so that in general
you search only very small regions. The areas defined by buckets can overlap. A document can exist in multiple
buckets.
A haystack index can reference two fields: the location field and a second field. The second field is used for exact
matches. Haystack indexes return documents based on location and an exact match on a single additional criterion.
These indexes are not necessarily suited to returning the closest documents to a particular location.
To build a haystack index, use the following syntax:
db.coll.ensureIndex( { <location field> : "geoHaystack" ,
<additional field> : 1 } ,
{ bucketSize : <bucket value> } )

Example
If you have a collection with documents that contain fields similar to the following:
{ _id : 100, pos: { lng : 126.9, lat : 35.2 } , type : "restaurant"}
{ _id : 200, pos: { lng : 127.5, lat : 36.1 } , type : "restaurant"}
{ _id : 300, pos: { lng : 128.0, lat : 36.7 } , type : "national park"}

The following operations create a haystack index with buckets that store keys within 1 unit of longitude or latitude.
db.places.ensureIndex( { pos : "geoHaystack", type : 1 } ,
{ bucketSize : 1 } )

This index stores the document with an _id field that has the value 200 in two different buckets:
In a bucket that includes the document where the _id field has a value of 100
In a bucket that includes the document where the _id field has a value of 300

7.3. Indexing Tutorials

355

MongoDB Documentation, Release 2.4.9

To query using a haystack index you use the geoSearch command. See Query a Haystack Index (page 356).
By default, queries that use a haystack index return 50 documents.
Query a Haystack Index
A haystack index is a special 2d geospatial index that is optimized to return results over small areas. To create a
haystack index see Create a Haystack Index (page 355).
To query a haystack index, use the geoSearch command. You must specify both the coordinates and the additional
field to geoSearch. For example, to return all documents with the value restaurant in the type field near the
example point, the command would resemble:
db.runCommand( { geoSearch : "places" ,
search : { type: "restaurant" } ,
near : [-74, 40.74] ,
maxDistance : 10 } )

Note: Haystack indexes are not suited to queries for the complete list of documents closest to a particular location.
The closest documents could be more distant compared to the bucket size.
Note: Spherical query operations (page 356) are not currently supported by haystack indexes.
The find() method and geoNear command cannot access the haystack index.

Calculate Distance Using Spherical Geometry

Note: While basic queries using spherical distance are supported by the 2d index, consider moving to a 2dsphere
index if your data is primarily longitude and latitude.
The 2d index supports queries that calculate distances on a Euclidean plane (flat surface). The index also supports the
following query operators and command that calculate distances using spherical geometry:
$nearSphere
$centerSphere
$near
geoNear command with the { spherical:

true } option.

Important: These three queries use radians for distance. Other query types do not.
For spherical query operators to function properly, you must convert distances to radians, and convert from radians to
the distances units used by your application.
To convert:
distance to radians: divide the distance by the radius of the sphere (e.g. the Earth) in the same units as the
distance measurement.
radians to distance: multiply the radian measure by the radius of the sphere (e.g. the Earth) in the units system
that you want to convert the distance to.

356

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

The radius of the Earth is approximately 3,959 miles or 6,371 kilometers.

The following query would return documents from the places collection within the circle described by the center [
-74, 40.74 ] with a radius of 100 miles:
db.places.find( { loc: { $geoWithin: { $centerSphere: [ [ -74, 40.74 ] ,
100 / 3959 ] } } } )

You may also use the distanceMultiplier option to the geoNear to convert radians in the mongod process,
rather than in your application code. See distance multiplier (page 357).
The following spherical query, returns all documents in the collection places within 100 miles from the point [
-74, 40.74 ].
db.runCommand( { geoNear: "places",
near: [ -74, 40.74 ],
spherical: true
} )

The output of the above command would be:

{
// [ ... ]
"results" : [
{
"dis" : 0.01853688938212826,
"obj" : {
"_id" : ObjectId( ... )
"loc" : [
-73,
40
]
}
}
],
"stats" : {
// [ ... ]
"avgDistance" : 0.01853688938212826,
"maxDistance" : 0.01853714811400047
},
"ok" : 1
}

Warning: Spherical queries that wrap around the poles or at the transition from -180 to 180 longitude raise an
error.
Note: While the default Earth-like bounds for geospatial indexes are between -180 inclusive, and 180, valid values
for latitude are between -90 and 90.

Distance Multiplier

The distanceMultiplier option of the geoNear command returns distances only after multiplying the results
by an assigned value. This allows MongoDB to return converted values, and removes the requirement to convert units
in application logic.

7.3. Indexing Tutorials

357

MongoDB Documentation, Release 2.4.9

Using distanceMultiplier in spherical queries provides results from the geoNear command that do not need
radian-to-distance conversion. The following example uses distanceMultiplier in the geoNear command
with a spherical (page 356) example:
db.runCommand( { geoNear: "places",
near: [ -74, 40.74 ],
spherical: true,
distanceMultiplier: 3959
} )

The output of the above operation would resemble the following:

{
// [ ... ]
"results" : [
{
"dis" : 73.46525170413567,
"obj" : {
"_id" : ObjectId( ... )
"loc" : [
-73,
40
]
}
}
],
"stats" : {
// [ ... ]
"avgDistance" : 0.01853688938212826,
"maxDistance" : 0.01853714811400047
},
"ok" : 1
}

7.3.4 Text Search Tutorials

Instructions for enabling MongoDBs text search feature, and for building and configuring text indexes.
Enable Text Search (page 359) You must explicitly enable text search in order to search string content in collections.
Create a text Index (page 359) A text index allows searches on text strings in the indexs specified fields.
Search String Content for Text (page 360) Use queries to find strings of text within collections.
Specify a Language for Text Index (page 363) The specified language determines the list of stop words and the rules
for Text Searchs stemmer and tokenizer.
Create text Index with Long Name (page 364) Override the text index name limit for long index names.
Control Search Results with Weights (page 365) Give priority to certain search values by denoting the significance
of an indexed field relative to other indexed fields
Limit the Number of Entries Scanned (page 366) Search only those documents that match a set of filter conditions.
Create text Index to Satisfy the filter Component of Text Search (page 367) Perform text searches that return results
without the need to scan documents.

358

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

Enable Text Search

New in version 2.4.
The text search (page 333) is currently a beta feature. As a beta feature:
You need to explicitly enable the feature before creating a text index (page 332) or using the text command.
To enable text search on replica sets (page 381) and sharded clusters (page 484), you need to enable on each
and every mongod for replica sets and on each and every mongos for sharded clusters.
Warning:
Do not enable or use text search on production systems.
Text indexes have significant storage requirements and performance costs. See Storage Requirements and
Performance Costs (page 333) for more information.
You can enable the text search feature at startup with the textSearchEnabled parameter:
mongod --setParameter textSearchEnabled=true

You may prefer to set the textSearchEnabled parameter in the configuration file.
Additionally, you can enable the feature in the mongo shell with the setParameter command. This command
does not propagate from the primary to the secondaries. You must enable on each and every mongod for replica sets.
Note: You must set the parameter every time you start the server. You may prefer to add the parameter to the
configuration files.

Create a text Index

You can create a text index on the field or fields whose value is a string or an array of string elements. When creating
a text index on multiple fields, you can specify the individual fields or you can wildcard specifier ($**).
Index Specific Fields

The following example creates a text index on the fields subject and content:
db.collection.ensureIndex(
{
subject: "text",
content: "text"
}
)

This text index catalogs all string data in the subject field and the content field, where the field value is either
a string or an array of string elements.
Index All Fields

To allow for text search on all fields with string content, use the wildcard specifier ($**) to index all fields that contain
string content.
The following example indexes any string value in the data of every field of every document in collection and
names the index TextIndex:

7.3. Indexing Tutorials

359

MongoDB Documentation, Release 2.4.9

db.collection.ensureIndex(
{ "$**": "text" },
{ name: "TextIndex" }
)

Search String Content for Text

In 2.4, you can enable the text search feature to create text indexes and issue text queries using the text.
The following tutorial offers various query patterns for using the text search feature.
The examples in this tutorial use a collection quotes that has a text index on the fields quote that contains a
string and related_quotes that contains an array of string elements.
Note: You cannot combine the text command, which requires a special text index (page 332), with a query operator
that requires a different type of special index. For example you cannot combine text with the $near operator.

Search for a Term

The following command searches for the word TOMORROW:

db.quotes.runCommand( "text", { search: "TOMORROW" } )

Because text command is case-insensitive, the text search will match the following document in the quotes collection:
{
"_id" : ObjectId("50ecef5f8abea0fda30ceab3"),
"quote" : "tomorrow, and tomorrow, and tomorrow, creeps in this petty pace",
"related_quotes" : [
"is this a dagger which I see before me",
"the handle toward my hand?"
],
"src" : {
"title" : "Macbeth",
"from" : "Act V, Scene V"
},
"speaker" : "macbeth"
}

Match Any of the Search Terms

If the search string is a space-delimited text, text command performs a logical OR search on each term and returns
documents that contains any of the terms.
For example, the search string "tomorrow largo" searches for the term tomorrow OR the term largo:
db.quotes.runCommand( "text", { search: "tomorrow largo" } )

The command will match the following documents in the quotes collection:
{
"_id" : ObjectId("50ecef5f8abea0fda30ceab3"),
"quote" : "tomorrow, and tomorrow, and tomorrow, creeps in this petty pace",
"related_quotes" : [

360

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

"is this a dagger which I see before me",

"the handle toward my hand?"
],
"src" : {
"title" : "Macbeth",
"from" : "Act V, Scene V"
},
"speaker" : "macbeth"
}
{
"_id" : ObjectId("50ecf0cd8abea0fda30ceab4"),
"quote" : "Es tan corto el amor y es tan largo el olvido.",
"related_quotes" : [
"Como para acercarla mi mirada la busca.",
"Mi corazn la busca, y ella no est conmigo."
],
"speaker" : "Pablo Neruda",
"src" : {
"title" : "Veinte poemas de amor y una cancin desesperada",
"from" : "Poema 20"
}
}

Match Phrases

To match the exact phrase that includes a space(s) as a single term, escape the quotes.
For example, the following command searches for the exact phrase "and tomorrow":
db.quotes.runCommand( "text", { search: "\"and tomorrow\"" } )

If the search string contains both phrases and individual terms, the text command performs a compound logical AND
of the phrases with the compound logical OR of the single terms, including the individual terms from each phrase.
For example, the following search string contains both individual terms corto and largo as well as the phrase
\"and tomorrow\":
db.quotes.runCommand( "text", { search: "corto largo \"and tomorrow\"" } )

The text command performs the equivalent to the following logical operation, where the individual terms corto,
largo, as well as the term tomorrow from the phrase "and tomorrow", are part of a logical OR expression:
(corto OR largo OR tomorrow) AND ("and tomorrow")

As such, the results for this search will include documents that only contain the phrase "and tomorrow" as well as
documents that contain the phrase "and tomorrow" and the terms corto and/or largo. Documents that contain
the phrase "and tomorrow" as well as the terms corto and largo will generally receive a higher score for this
search.
Match Some Words But Not Others

A negated term is a term that is prefixed by a minus sign -. If you negate a term, the text command will exclude the
documents that contain those terms from the results.
Note: If the search text contains only negated terms, the text command will not return any results.

7.3. Indexing Tutorials

361

MongoDB Documentation, Release 2.4.9

The following example returns those documents that contain the term tomorrow but not the term petty.
db.quotes.runCommand( "text" , { search: "tomorrow -petty" } )

Limit the Number of Matching Documents in the Result Set

Note: The result from the text command must fit within the maximum BSON Document Size.
By default, the text command will return up to 100 matching documents, from highest to lowest scores. To override
this default limit, use the limit option in the text command, as in the following example:
db.quotes.runCommand( "text", { search: "tomorrow", limit: 2 } )

The text command will return at most 2 of the highest scoring results.
The limit can be any number as long as the result set fits within the maximum BSON Document Size.
Specify Which Fields to Return in the Result Set

In the text command, use the project option to specify the fields to include (1) or exclude (0) in the matching
documents.
Note: The _id field is always returned unless explicitly excluded in the project document.
The following example returns only the _id field and the src field in the matching documents:
db.quotes.runCommand( "text", { search: "tomorrow",
project: { "src": 1 } } )

Search with Additional Query Conditions

The text command can also use the filter option to specify additional query conditions.
The following example will return the documents that contain the term tomorrow AND the speaker is macbeth:
db.quotes.runCommand( "text", { search: "tomorrow",
filter: { "speaker" : "macbeth" } } )

See also:
Limit the Number of Entries Scanned (page 366)
Search for Text in Specific Languages

You can specify the language that determines the tokenization, stemming, and removal of stop words, as in the following example:
db.quotes.runCommand( "text", { search: "amor", language: "spanish" } )

See text-search-languages for a list of supported languages as well as Specify a Language for Text Index (page 363)
for specifying languages for the text index.

362

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

Text Search Output

The text command returns a document that contains the result set.
See text-search-output for information on the output.
Specify a Language for Text Index
This tutorial describes how to specify the default language associated with the text index (page 363) and also how to
create text indexes for collections that contain documents in different languages (page 363).
Specify the Default Language for a text Index

The default language associated with the indexed data determines the list of stop words and the rules for the stemmer
and tokenizer. The default language for the indexed data is english.
To specify a different language, use the default_language option when creating the text index. See textsearch-languages for the languages available for default_language.
The following example creates a text index on the content field and sets the default_language to spanish:
db.collection.ensureIndex(
{ content : "text" },
{ default_language: "spanish" }
)

Create a text Index for a Collection in Multiple Languages

Specify the Index Language within the Document If a collection contains documents that are in different languages, include a field in the documents that contain the language to use:
If you include a field named language in the document, by default, the ensureIndex() method will use
the value of this field to override the default language.
To use a field with a name other than language, you must specify the name of this field to the
ensureIndex() method with the language_override option.
See text-search-languages for a list of supported languages.
Include the language Field Include a field language that specifies the language to use for the individual documents.
For example, the documents of a multi-language collection quotes contain the field language:
{ _id: 1, language: "portuguese", quote: "A sorte protege os audazes" }
{ _id: 2, language: "spanish", quote: "Nada hay ms surreal que la realidad." }
{ _id: 3, language: "english", quote: "is this a dagger which I see before me" }

Create a text index on the field quote:

db.quotes.ensureIndex( { quote: "text" } )

For the documents that contain the language field, the text index uses that language to determine the stop
words and the rules for the stemmer and the tokenizer.

7.3. Indexing Tutorials

363

MongoDB Documentation, Release 2.4.9

For documents that do not contain the language field, the index uses the default language, which is English,
to determine the stop words and rules for the stemmer and the tokenizer.
For example, the Spanish word que is a stop word. So the following text command would not match any document:
db.quotes.runCommand( "text", { search: "que", language: "spanish" } )

Use any Field to Specify the Language for a Document Include a field that specifies the language to use for the
individual documents. To use a field with a name other than language, include the language_override option
when creating the index.
For example, the documents of a multi-language collection quotes contain the field idioma:
{ _id: 1, idioma: "portuguese", quote: "A sorte protege os audazes" }
{ _id: 2, idioma: "spanish", quote: "Nada hay ms surreal que la realidad." }
{ _id: 3, idioma: "english", quote: "is this a dagger which I see before me" }

Create a text index on the field quote with the language_override option:
db.quotes.ensureIndex( { quote : "text" },
{ language_override: "idioma" } )

For the documents that contain the idioma field, the text index uses that language to determine the stop
words and the rules for the stemmer and the tokenizer.
For documents that do not contain the idioma field, the index uses the default language, which is English, to
determine the stop words and rules for the stemmer and the tokenizer.
For example, the Spanish word que is a stop word. So the following text command would not match any document:
db.quotes.runCommand( "text", { search: "que", language: "spanish" } )

Create text Index with Long Name

The default name for the index consists of each indexed field name concatenated with _text. For example, the
following command creates a text index on the fields content, users.comments, and users.profiles:
db.collection.ensureIndex(
{
content: "text",
"users.comments": "text",
"users.profiles": "text"
}
)

The default name for the index is:

"content_text_users.comments_text_users.profiles_text"

To avoid creating an index with a name that exceeds the index name length limit, you can pass the name
option to the db.collection.ensureIndex() method:
db.collection.ensureIndex(
{
content: "text",
"users.comments": "text",
"users.profiles": "text"
},
{

364

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

name: "MyTextIndex"
}
)

Note:
To drop the text index, use the index name.
db.collection.getIndexes().

To get the name of an index, use

Control Search Results with Weights

This document describes how to create a text index with specified weights for results fields.
By default, the text command returns matching documents based on scores, from highest to lowest. For a text
index, the weight of an indexed field denotes the significance of the field relative to the other indexed fields in terms
of the score. The score for a given word in a document is derived from the weighted sum of the frequency for each of
the indexed fields in that document.
The default weight is 1 for the indexed fields. To adjust the weights for the indexed fields, include the weights
option in the db.collection.ensureIndex() method.
Warning: Choose the weights carefully in order to prevent the need to reindex.
A collection blog has the following documents:
{ _id: 1,
content: "This morning I had a cup of coffee.",
about: "beverage",
keywords: [ "coffee" ]
}
{ _id: 2,
content: "Who doesn't like cake?",
about: "food",
keywords: [ "cake", "food", "dessert" ]
}

To create a text index with different field weights for the content field and the keywords field, include the
weights option to the ensureIndex() method. For example, the following command creates an index on three
fields and assigns weights to two of the fields:
db.blog.ensureIndex(
{
content: "text",
keywords: "text",
about: "text"
},
{
weights: {
content: 10,
keywords: 5,
},
name: "TextIndex"
}
)

The text index has the following fields and weights:

7.3. Indexing Tutorials

365

MongoDB Documentation, Release 2.4.9

content has a weight of 10,

keywords has a weight of 5, and
about has the default weight of 1.
These weights denote the relative significance of the indexed fields to each other. For instance, a term match in the
content field has:
2 times (i.e. 10:5) the impact as a term match in the keywords field and
10 times (i.e. 10:1) the impact as a term match in the about field.
Limit the Number of Entries Scanned
This tutorial describes how to limit the text search to scan only those documents with a field value.
The text command includes the filter option to further restrict the results of a text search. For a filter that
specifies equality conditions, this tutorial demonstrates how to perform text searches on only those documents that
match the filter conditions, as opposed to performing a text search first on all the documents and then matching
on the filter condition.
Consider a collection inventory that contains the following documents:
{
{
{
{
{
{

_id:
_id:
_id:
_id:
_id:
_id:

1,
2,
3,
4,
5,
6,

dept:
dept:
dept:
dept:
dept:
dept:

"tech", description: "a

"tech", description: "a
"kitchen", description:
"kitchen", description:
"food", description: "a
"food", description: "a

fun green computer" }

wireless red mouse" }
"a green placemat" }
"a red peeler" }
green apple" }
red potato" }

A common use case is to perform text searches by individual departments, such as:
db.inventory.runCommand( "text", {
search: "green",
filter: { dept : "kitchen" }
}
)

To limit the text search to scan only those documents within a specific dept, create a compound index that specifies
an ascending/descending index key on the field dept and a text index key on the field description:
db.inventory.ensureIndex(
{
dept: 1,
description: "text"
}
)

Important:
The ascending/descending index keys must be listed before, or prefix, the text index keys.
By prefixing the text index fields with ascending/descending index fields, MongoDB will only index documents that have the prefix fields.
You cannot include multi-key (page 324) index fields or geospatial (page 327) index fields.
The text command must include the filter option that specifies an equality condition for the prefix fields.

366

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

Then, the text search within a particular department will limit the scan of indexed documents. For example, the
following text command scans only those documents with dept equal to kitchen:
db.inventory.runCommand( "text", {
search: "green",
filter: { dept : "kitchen" }
}
)

The returned result includes the statistics that shows that the command scanned 1 document, as indicated by the
nscanned field:
{
"queryDebugString" : "green||||||",
"language" : "english",
"results" : [
{
"score" : 0.75,
"obj" : {
"_id" : 3,
"dept" : "kitchen",
"description" : "a green placemat"
}
}
],
"stats" : {
"nscanned" : 1,
"nscannedObjects" : 0,
"n" : 1,
"nfound" : 1,
"timeMicros" : 211
},
"ok" : 1
}

For more information on the result set, see text-search-output.

Create text Index to Satisfy the filter Component of Text Search
To perform a text search that also includes additional query conditions, the text command provides a filter
option. A compound index can support the text search and the filter condition, such that the operation does
not need to scan any documents to satisfy the filter condition; i.e. the returned text.stats contain 0 for the
nscannedObjects field.
Procedure

To create a text index that can fulfill the filter condition of a text search:
1. Append scalar index fields to a text index, as in the following example which specifies an ascending index
key on cited:
db.quotes.ensureIndex(
{
comments: "text",
cited: 1

7.3. Indexing Tutorials

367

MongoDB Documentation, Release 2.4.9

}
)

2. Use the filter option in the text to specify an condition on the cited field, as in the following:
db.quotes.runCommand( "text",
{
search: "tomorrow",
filter: { cited: { $gt: 10 } }
}
)

Considerations

When creating a compound index with that includes the text index, you cannot include multi-key (page 324) index
field or geospatial (page 327) index field.
With a compound index that includes the text index and an ascending/descending key or keys, sort operations do not
use the ascending/descending key from this index; only the text score determines the sort order.

7.3.5 Indexing Strategies

The best indexes for your application must take a number of factors into account, including the kinds of queries you
expect, the ratio of reads to writes, and the amount of free memory on your system.
When developing your indexing strategy you should have a deep understanding of your applications queries. Before
you build indexes, map out the types of queries you will run so that you can build indexes that reference those fields.
Indexes come with a performance cost, but are more than worth the cost for frequent queries on large data set. Consider
the relative frequency of each query in the application and whether the query justifies an index.
The best overall strategy for designing indexes is to profile a variety of index configurations with data sets similar to
the ones youll be running in production to see which configurations perform best.Inspect the current indexes created
for your collections to ensure they are supporting your current and planned queries. If an index is no longer used, drop
the index.
MongoDB can only use one index to support any given operation. However, each clause of an $or query may use a
different index.
The following documents introduce indexing strategies:
Create Indexes to Support Your Queries (page 368) An index supports a query when the index contains all the fields
scanned by the query. Creating indexes that supports queries results in greatly increased query performance.
Use Indexes to Sort Query Results (page 371) To support efficient queries, use the strategies here when you specify
the sequential order and sort order of index fields.
Ensure Indexes Fit in RAM (page 373) When your index fits in RAM, the system can avoid reading the index from
disk and you get the fastest processing.
Create Queries that Ensure Selectivity (page 373) Selectivity is the ability of a query to narrow results using the
index. Selectivity allows MongoDB to use the index for a larger portion of the work associated with fulfilling
the query.
Create Indexes to Support Your Queries
An index supports a query when the index contains all the fields scanned by the query. The query scans the index and
not the collection. Creating indexes that supports queries results in greatly increased query performance.
368

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

This document describes strategies for creating indexes that support queries.
Create a Single-Key Index if All Queries Use the Same, Single Key

If you only ever query on a single key in a given collection, then you need to create just one single-key index for that
collection. For example, you might create an index on category in the product collection:
db.products.ensureIndex( { "category": 1 } )

Create Compound Indexes to Support Several Different Queries

If you sometimes query on only one key and at other times query on that key combined with a second key, then creating
a compound index is more efficient than creating a single-key index. MongoDB will use the compound index for both
queries. For example, you might create an index on both category and item.
db.products.ensureIndex( { "category": 1, "item": 1 } )

This allows you both options. You can query on just category, and you also can query on category combined
with item. A single compound index (page 322) on multiple fields can support all the queries that search a prefix
subset of those fields.
Note: With the exception of queries that use the $or operator, a query does not use multiple indexes. A query uses
only one index.
Example
The following index on a collection:
{ x: 1, y: 1, z: 1 }

Can support queries that the following indexes support:

{ x: 1 }
{ x: 1, y: 1 }

There are some situations where the prefix indexes may offer better query performance: for example if z is a large
array.
The { x:

1, y:

1, z:

1 } index can also support many of the same queries as the following index:

{ x: 1, z: 1 }

Also, { x:

1, z:

1 } has an additional use. Given the following query:

db.collection.find( { x: 5 } ).sort( { z: 1} )

The { x: 1, z: 1 } index supports both the query and the sort operation, while the { x: 1, y: 1,
z: 1 } index only supports the query. For more information on sorting, see Use Indexes to Sort Query Results
(page 371).

Create Indexes that Support Covered Queries

A covered query is a query in which:

all the fields in the query (page 60) are part of an index, and
7.3. Indexing Tutorials

369

MongoDB Documentation, Release 2.4.9

all the fields returned in the results are in the same index.
Because the index covers the query, MongoDB can both match the query conditions (page 60) and return the results
using only the index; MongoDB does not need to look at the documents, only the index, to fulfill the query. An index
can also cover an aggregation pipeline operation (page 283) on unsharded collections.
Querying only the index can be much faster than querying documents outside of the index. Index keys are typically
smaller than the documents they catalog, and indexes are typically available in RAM or located sequentially on disk.
MongoDB automatically uses an index that covers a query when possible. To ensure that an index can cover a query,
create an index that includes all the fields listed in the query document (page 60) and in the query result. You can
specify the fields to return in the query results with a projection (page 64) document. By default, MongoDB includes
the _id field in the query result. So, if the index does not include the _id field, then you must exclude the _id field
(i.e. _id: 0) from the query results.
Example
Given collection users with an index on the fields user and status, as created by the following option:
db.users.ensureIndex( { status: 1, user: 1 } )

Then, this index will cover the following query which selects on the status field and returns only the user field:
db.users.find( { status: "A" }, { user: 1, _id: 0 } )

In the operation, the projection document explicitly specifies _id:

the index is only on the status and the user fields.

0 to exclude the _id field from the result since

If the projection document does not specify the exclusion of the _id field, the query returns the _id field. The
following query is not covered by the index on the status and the user fields because with the projection document
{ user: 1 }, the query returns both the user field and the _id field:
db.users.find( { status: "A" }, { user: 1 } )

An index cannot cover a query if:

any of the indexed fields in any of the documents in the collection includes an array. If an indexed field is an
array, the index becomes a multi-key index (page 324) index and cannot support a covered query.
any of the indexed fields are fields in subdocuments. To index fields in subdocuments, use dot notation. For
example, consider a collection users with documents of the following form:
{ _id: 1, user: { login: "tester" } }

The collection has the following indexes:

{ user: 1 }
{ "user.login": 1 }

The { user:

1 } index covers the following query:

db.users.find( { user: { login: "tester" } }, { user: 1, _id: 0 } )

However, the { "user.login":

1 } index does not cover the following query:

db.users.find( { "user.login": "tester" }, { "user.login": 1, _id: 0 } )

The query, however, does use the { "user.login":

370

1 } index to find matching documents.

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

To determine whether a query is a covered query, use the explain() method. If the explain() output displays
true for the indexOnly field, the query is covered by an index, and MongoDB queries only that index to match
the query and return the results.
For more information see Measure Index Use (page 349).
Use Indexes to Sort Query Results
In MongoDB sort operations that sort documents based on an indexed field provide the greatest performance. Indexes
in MongoDB, as in other databases, have an order: as a result, using an index to access documents returns in the same
order as the index.
To sort on multiple fields, create a compound index (page 322). With compound indexes, the results can be in the
sorted order of either the full index or an index prefix. An index prefix is a subset of a compound index; the subset
consists of one or more fields at the start of the index, in order. For example, given an index { a:1, b: 1, c:
1, d: 1 }, the following subsets are index prefixes:
{ a: 1 }
{ a: 1, b: 1 }
{ a: 1, b: 1, c: 1 }

For more information on sorting by index prefixes, see Sort Subset Starts at the Index Beginning (page 372).
If the query includes equality match conditions on an index prefix, you can sort on a subset of the index that starts
after or overlaps with the prefix. For example, given an index { a: 1, b: 1, c: 1, d: 1 }, if the
query condition includes equality match conditions on a and b, you can specify a sort on the subsets { c: 1 } or
{ c: 1, d: 1 }:
db.collection.find( { a: 5, b: 3 } ).sort( { c: 1 } )
db.collection.find( { a: 5, b: 3 } ).sort( { c: 1, d: 1 } )

In these operations, the equality match and the sort documents together cover the index prefixes { a:
c: 1 } and { a: 1, b: 1, c: 1, d: 1 } respectively.

1, b:

1,

You can also specify a sort order that includes the prefix; however, since the query condition specifies equality matches
on these fields, they are constant in the resulting documents and do not contribute to the sort order:
db.collection.find( { a: 5, b: 3 } ).sort( { a: 1, b: 1, c: 1 } )
db.collection.find( { a: 5, b: 3 } ).sort( { a: 1, b: 1, c: 1, d: 1 } )

For more information on sorting by index subsets that are not prefixes, see Sort Subset Does Not Start at the Index
Beginning (page 372).
Note: For in-memory sorts that do not use an index, the sort() operation is significantly slower. The sort()
operation will abort when it uses 32 megabytes of memory.

Sort With a Subset of Compound Index

If the sort document contains a subset of the compound index fields, the subset can determine whether MongoDB can
use the index efficiently to both retrieve and sort the query results. If MongoDB can efficiently use the index to both
retrieve and sort the query results, the output from the explain() will display scanAndOrder as false or 0.
If MongoDB can only use the index for retrieving documents that meet the query criteria, MongoDB must manually
sort the resulting documents without the use of the index. For in-memory sort operations, explain() will display
scanAndOrder as true or 1.

7.3. Indexing Tutorials

371

MongoDB Documentation, Release 2.4.9

Sort Subset Starts at the Index Beginning If the sort document is a subset of a compound index and starts from
the beginning of the index, MongoDB can use the index to both retrieve and sort the query results.
For example, the collection collection has the following index:
{ a: 1, b: 1, c: 1, d: 1 }

The following operations include a sort with a subset of the index. Because the sort subset starts at beginning of the
index, the operations can use the index for both the query retrieval and sort:
db.collection.find().sort( { a:1 } )
db.collection.find().sort( { a:1, b:1 } )
db.collection.find().sort( { a:1, b:1, c:1 } )
db.collection.find( { a: 4 } ).sort( { a: 1, b: 1 } )
db.collection.find( { a: { $gt: 4 } } ).sort( { a: 1, b: 1 } )
db.collection.find( { b: 5 } ).sort( { a: 1, b: 1 } )
db.collection.find( { b: { $gt:5 }, c: { $gt: 1 } } ).sort( { a: 1, b: 1 } )

The last two operations include query conditions on the field b but does not include a query condition on the field a:
db.collection.find( { b: 5 } ).sort( { a: 1, b: 1 } )
db.collection.find( { b: { $gt:5 }, c: { $gt: 1 } } ).sort( { a: 1, b: 1 } )

Consider the case where the collection has the index { b: 1 } in addition to the { a: 1, b: 1, c: 1,
d: 1 } index. Because of the query condition on b, it is not immediately obvious which index MongoDB may
select as the best index. To explicitly specify the index to use, see hint().
Sort Subset Does Not Start at the Index Beginning The sort document can be a subset of a compound index that
does not start from the beginning of the index. For instance, { c: 1 } is a subset of the index { a: 1, b:
1, c: 1, d: 1 } that omits the preceding index fields a and b. MongoDB can use the index efficiently if the
query document includes all the preceding fields of the index, in this case a and b, in equality conditions. In other
words, the equality conditions in the query document and the subset in the sort document contiguously cover a prefix
of the index.
For example, the collection collection has the following index:
{ a: 1, b: 1, c: 1, d: 1 }

Then following operations can use the index efficiently:

db.collection.find( { a: 5 } ).sort( { b: 1, c: 1 } )
db.collection.find( { a: 5, c: 4, b: 3 } ).sort( { d: 1 } )

In the first operation, the query document { a: 5 } with the sort document { b:
the prefix { a:1 , b: 1, c: 1 } of the index.
In the second operation, the query document { a:
1 } covers the full index.

5, c:

4, b:

1, c:

1 } cover

3 } with the sort document { d:

Only the index fields preceding the sort subset must have the equality conditions in the query document. The other
index fields may have other conditions. The following operations can efficiently use the index since the equality
conditions in the query document and the subset in the sort document contiguously cover a prefix of the index:
db.collection.find( { a: 5, b: 3 } ).sort( { c: 1 } )
db.collection.find( { a: 5, b: 3, c: { $lt: 4 } } ).sort( { c: 1 } )

The following operations specify a sort document of { c:

matches on the preceding index fields a and b:

372

1 }, but the query documents do not contain equality

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

db.collection.find( { a: { $gt: 2 } } ).sort( { c: 1 } )

db.collection.find( { c: 5 } ).sort( { c: 1 } )

These operations will not efficiently use the index { a:

the index to retrieve the documents.

1, b:

1, c:

1, d:

1 } and may not even use

Ensure Indexes Fit in RAM

For the fastest processing, ensure that your indexes fit entirely in RAM so that the system can avoid reading the index
from disk.
To check the size of your indexes, use the db.collection.totalIndexSize() helper, which returns data in
bytes:
> db.collection.totalIndexSize()
4294976499

The above example shows an index size of almost 4.3 gigabytes. To ensure this index fits in RAM, you must not only
have more than that much RAM available but also must have RAM available for the rest of the working set. Also
remember:
If you have and use multiple collections, you must consider the size of all indexes on all collections. The indexes and
the working set must be able to fit in memory at the same time.
There are some limited cases where indexes do not need to fit in memory. See Indexes that Hold Only Recent Values
in RAM (page 373).
See also:
collStats and db.collection.stats()
Indexes that Hold Only Recent Values in RAM

Indexes do not have to fit entirely into RAM in all cases. If the value of the indexed field increments with every insert,
and most queries select recently added documents; then MongoDB only needs to keep the parts of the index that hold
the most recent or right-most values in RAM. This allows for efficient index use for read and write operations and
minimize the amount of RAM required to support the index.
Create Queries that Ensure Selectivity
Selectivity is the ability of a query to narrow results using the index. Effective indexes are more selective and allow
MongoDB to use the index for a larger portion of the work associated with fulfilling the query.
To ensure selectivity, write queries that limit the number of possible documents with the indexed field. Write queries
that are appropriately selective relative to your indexed data.
Example
Suppose you have a field called status where the possible values are new and processed. If you add an index
on status youve created a low-selectivity index. The index will be of little help in locating records.
A better strategy, depending on your queries, would be to create a compound index (page 322) that includes the lowselectivity field and another field. For example, you could create a compound index on status and created_at.
Another option, again depending on your use case, might be to use separate collections, one for each status.

7.3. Indexing Tutorials

373

MongoDB Documentation, Release 2.4.9

Example
Consider an index { a : 1 } (i.e. an index on the key a sorted in ascending order) on a collection where a has
three values evenly distributed across the collection:
{
{
{
{
{
{
{
{
{

_id:
_id:
_id:
_id:
_id:
_id:
_id:
_id:
_id:

ObjectId(),
ObjectId(),
ObjectId(),
ObjectId(),
ObjectId(),
ObjectId(),
ObjectId(),
ObjectId(),
ObjectId(),

a:
a:
a:
a:
a:
a:
a:
a:
a:

1,
1,
1,
2,
2,
2,
3,
3,
3,

b:
b:
b:
b:
b:
b:
b:
b:
b:

"ab"
"cd"
"ef"
"jk"
"lm"
"no"
"pq"
"rs"
"tv"

}
}
}
}
}
}
}
}
}

If you query for { a: 2, b: "no" } MongoDB must scan 3 documents in the collection to return the one
matching result. Similarly, a query for { a: { $gt: 1}, b: "tv" } must scan 6 documents, also to
return one result.
Consider the same index on a collection where a has nine values evenly distributed across the collection:
{
{
{
{
{
{
{
{
{

_id:
_id:
_id:
_id:
_id:
_id:
_id:
_id:
_id:

ObjectId(),
ObjectId(),
ObjectId(),
ObjectId(),
ObjectId(),
ObjectId(),
ObjectId(),
ObjectId(),
ObjectId(),

a:
a:
a:
a:
a:
a:
a:
a:
a:

1,
2,
3,
4,
5,
6,
7,
8,
9,

b:
b:
b:
b:
b:
b:
b:
b:
b:

"ab"
"cd"
"ef"
"jk"
"lm"
"no"
"pq"
"rs"
"tv"

}
}
}
}
}
}
}
}
}

If you query for { a: 2, b: "cd" }, MongoDB must scan only one document to fulfill the query. The index
and query are more selective because the values of a are evenly distributed and the query can select a specific document
using the index.
However, although the index on a is more selective, a query such as { a:
still need to scan 4 documents.

{ $gt:

5 }, b:

"tv" } would

If overall selectivity is low, and if MongoDB must read a number of documents to return results, then some queries
may perform faster without indexes. To determine performance, see Measure Index Use (page 349).
For a conceptual introduction to indexes in MongoDB see Index Concepts (page 318).

374

Chapter 7. Indexes

MongoDB Documentation, Release 2.4.9

7.4 Indexing Reference

7.4.1 Indexing Methods in the mongo Shell
Name
Description
db.collection.createIndex()
Builds an index on a collection. Use db.collection.ensureIndex().
db.collection.dropIndex()
Removes a specified index on a collection.
db.collection.dropIndexes()
Removes all indexes on a collection.
db.collection.ensureIndex()
Creates an index if it does not currently exist. If the index exists ensureIndex()
does nothing.
db.collection.getIndexes()
Returns an array of documents that describe the existing indexes on a collection.
db.collection.getIndexStats()
Renders a human-readable view of the data collected by indexStats which
reflects B-tree utilization.
db.collection.indexStats()
Renders a human-readable view of the data collected by indexStats which
reflects B-tree utilization.
db.collection.reIndex()
Rebuilds all existing indexes on a collection.
db.collection.totalIndexSize()
Reports the total size used by the indexes on a collection. Provides a wrapper around
the totalIndexSize field of the collStats output.
cursor.explain()
Reports on the query execution plan, including index use, for a cursor.
cursor.hint()
Forces MongoDB to use a specific index for a query.
cursor.max()
Specifies an exclusive upper index bound for a cursor. For use with
cursor.hint()
cursor.min()
Specifies an inclusive lower index bound for a cursor. For use with
cursor.hint()
cursor.snapshot() Forces the cursor to use the index on the _id field. Ensures that the cursor returns
each document, with regards to the value of the _id field, only once.

7.4.2 Indexing Database Commands

Name
dropIndexes
compact
reIndex
validate
indexStats
geoNear
geoSearch
geoWalk
checkShardingIndex

Description
Removes indexes from a collection.
Defragments a collection and rebuilds the indexes.
Rebuilds all indexes on a collection.
Internal command that scans for a collections data and indexes for correctness.
Experimental command that collects and aggregates statistics on all indexes.
Performs a geospatial query that returns the documents closest to a given point.
Performs a geospatial query that uses MongoDBs haystack index functionality.
An internal command to support geospatial queries.
Internal command that validates index on shard key.

7.4.3 Geospatial Query Selectors

Name
$geoWithin
$geoIntersects
$near
$nearSphere

Description
Selects geometries within a bounding GeoJSON geometry.
Selects geometries that intersect with a GeoJSON geometry.
Returns geospatial objects in proximity to a point.
Returns geospatial objects in proximity to a point on a sphere.

7.4. Indexing Reference

375

MongoDB Documentation, Release 2.4.9

7.4.4 Indexing Query Modifiers

Name
$explain
$hint
$max
$min
$returnKey
$snapshot

376

Description
Forces MongoDB to report on query execution plans. See explain().
Forces MongoDB to use a specific index. See hint()
Specifies an exclusive upper limit for the index to use in a query. See max().
Specifies an inclusive lower limit for the index to use in a query. See min().
Forces the cursor to only return fields included in the index.
Forces the query to use the index on the _id field. See snapshot().

Chapter 7. Indexes

CHAPTER 8

Replication

A replica set in MongoDB is a group of mongod processes that maintain the same data set. Replica sets provide
redundancy and high availability, and are the basis for all production deployments. This section introduces replication
in MongoDB as well as the components and architecture of replica sets. The section also provides tutorials for common
tasks related to replica sets.
Replication Introduction (page 377) An introduction to replica sets, their behavior, operation, and use.
Replication Concepts (page 381) The core documentation of replica set operations, configurations, architectures and
behaviors.
Replica Set Members (page 382) Introduces the components of replica sets.
Replica Set Deployment Architectures (page 390) Introduces architectural considerations related to replica
sets deployment planning.
Replica Set High Availability (page 397) Presents the details of the automatic failover and recovery process
with replica sets.
Replica Set Read and Write Semantics (page 402) Presents the semantics for targeting read and write operations to the replica set, with an awareness of location and set configuration.
Replica Set Tutorials (page 419) Tutorials for common tasks related to the use and maintenance of replica sets.
Replication Reference (page 466) Reference for functions and operations related to replica sets.

8.1 Replication Introduction

Replication is the process of synchronizing data across multiple servers.

8.1.1 Purposes of Replication

Replication provides redundancy and increases data availability. With multiple copies of data on different database
servers, replication protects a database from the loss of a single server. Replication also allows you to recover from
hardware failure and service interruptions. With additional copies of the data, you can dedicate a server to disaster
recovery, reporting, or backup.
In some cases, you can use replication to increase read capacity. Clients have the ability to send read operations to
different servers. You can also maintain copies in different data centers to increase the locality and availability of data
for distributed applications.

377

MongoDB Documentation, Release 2.4.9

8.1.2 Replication in MongoDB

A replica set is a group of mongod instances that host the same data set. One mongod, the primary, receives all write
operations. All other instances, secondaries, apply operations from the primary so that they have the same data set.
The primary accepts all write operations from clients. Replica sets can have one and only one primary at any given
moment. Because only one member can accept write operations, replica sets provide strict consistency. To support
replication, the primary records all changes to its data sets in its oplog (page 410). See primary (page 382) for more
information.

Figure 8.1: Diagram of default routing of reads and writes to the primary.
The secondaries replicate the primarys oplog and apply the operations to their data sets. Secondaries data sets reflect
the primarys data set. If the primary is unavailable, the replica set will elect a secondary to be primary. By default,
clients read from the primary; however, clients can specify a read preference (page 406) to send read operations to
secondaries. See secondaries (page 382) for more information.
You may add an extra mongod instance to a replica set as an arbiter. Arbiters do not maintain a data set. Arbiters
only exist to vote in elections. If your replica set has an even number of members, add an arbiter to obtain a majority
of votes in an election for primary. Arbiters do not require dedicated hardware. See arbiter (page 389) for more
information.

378

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Figure 8.2: Diagram of a 3 member replica set that consists of a primary and two secondaries.

Figure 8.3: Diagram of a replica set that consists of a primary, a secondary, and an arbiter.

8.1. Replication Introduction

379

MongoDB Documentation, Release 2.4.9

Note: The primary may, under some conditions, step down and become a secondary. A secondary may become
the primary during an election. An arbiter, however, will never change state and will always be an arbiter.

Asynchronous Replication
Secondaries apply operations from the primary asynchronously. By applying operations after the primary, replica sets
can continue to function without some members. However, as a result, secondaries may not return the most current
data to clients.
See Replica Set Oplog (page 410) and Replica Set Data Synchronization (page 412) for more information. See Read
Preference (page 406) for more on read operations and secondaries.
Automatic Failover
When a primary does not communicate with the other members of the replica set for more than 10 seconds, the replica
set will attempt to select another member to become the new primary. The first secondary that receives a majority of
the votes becomes primary.

Figure 8.4: Diagram of an election of a new primary. In a three member replica set with two secondaries, the primary
becomes unreachable. The loss of a primary triggers an election where one of the secondaries becomes the new
primary

380

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

See Replica Set Elections (page 397) and Rollbacks During Replica Set Failover (page 401) for more information.
Additional Features
Replica sets provide a number of options to support application needs. For example, you may deploy a replica set
with members in multiple data centers (page 396), or control the outcome of elections by adjusting the priority
(page 469) of some members. Replica sets also support configuring dedicated members for reporting, disaster recovery,
or backup functions.
See Priority 0 Replica Set Members (page 386), Hidden Replica Set Members (page 387) and Delayed Replica Set
Members (page 388) for more information.

8.2 Replication Concepts

These documents describe and provide examples of replica set operation, configuration, and behavior. For an overview
of replication, see Replication Introduction (page 377). For documentation of the administration of replica sets, see
Replica Set Tutorials (page 419). The Replication Reference (page 466) documents commands and operations specific
to replica sets.
Replica Set Members (page 382) Introduces the components of replica sets.
Replica Set Primary (page 382) The primary is the only member of a replica set that accepts write operations.
Replica Set Secondary Members (page 382) Secondary members replicate the primarys data set and accept
read operations. If the set has no primary, a secondary can become primary.
Priority 0 Replica Set Members (page 386) Priority 0 members are secondaries that cannot become the primary.
Hidden Replica Set Members (page 387) Hidden members are secondaries that are invisible to applications.
These members support dedicated workloads, such as reporting or backup.
Replica Set Arbiter (page 389) An arbiter does not maintain a copy of the data set but participate in elections.
Replica Set Deployment Architectures (page 390) Introduces architectural considerations related to replica sets deployment planning.
Three Member Replica Sets (page 392) Three-member replica sets provide the minimum recommended architecture for a replica set.
Replica Sets with Four or More Members (page 395) Four or more member replica sets provide greater redundancy and can support greater distribution of read operations and dedicated functionality.
Replica Set High Availability (page 397) Presents the details of the automatic failover and recovery process with
replica sets.
Replica Set Elections (page 397) Elections occur when the primary becomes unavailable and the replica set
members autonomously select a new primary.
Read Preference (page 406) Applications specify read preference to control how drivers direct read operations
to members of the replica set.
Replication Processes (page 410) Mechanics of the replication process and related topics.
Master Slave Replication (page 413) Master-slave replication provided redundancy in early versions of MongoDB.
Replica sets replace master-slave for most use cases.

8.2. Replication Concepts

381

MongoDB Documentation, Release 2.4.9

8.2.1 Replica Set Members

A replica set in MongoDB is a group of mongod processes that provide redundancy and high availability. The
members of a replica set are:
Primary (page ??). The primary receives all write operations.
Secondaries (page ??). Secondaries replicate operations from the primary to maintain an identical data set. Secondaries may have additional configurations for special usage profiles. For example, secondaries may be nonvoting (page 400) or priority 0 (page 386).
You can also maintain an arbiter (page ??) as part of a replica set. Arbiters do not keep a copy of the data. However,
arbiters play a role in the elections that select a primary if the current primary is unavailable.
A replica set can have up to 12 members.

However, only 7 members can vote at a time.

The minimum requirements for a replica set are: A primary (page ??), a secondary (page ??), and an arbiter (page ??).
Most deployments, however, will keep three members that store data: A primary (page ??) and two secondary members
(page ??).
Replica Set Primary
The primary is the only member in the replica set that receives write operations. MongoDB applies write operations
on the primary and then records the operations on the primarys oplog (page 410). Secondary (page ??) members
replicate this log and apply the operations to their data sets.
In the following three-member replica set, the primary accepts all write operations. Then the secondaries replicate the
oplog to apply to their data sets.
All members of the replica set can accept read operations. However, by default, an application directs its read operations to the primary member. See Read Preference (page 406) for details on changing the default read behavior.
The replica set can have at most one primary. If the current primary becomes unavailable, an election determines the
new primary. See Replica Set Elections (page 397) for more details.
In the following 3-member replica set, the primary becomes unavailable. This triggers an election which selects one
of the remaining secondaries as the new primary.
Replica Set Secondary Members
A secondary maintains a copy of the primarys data set. To replicate data, a secondary applies operations from the
primarys oplog (page 410) to its own data set in an asynchronous process. A replica set can have one or more
secondaries.
The following three-member replica set has two secondary members. The secondaries replicate the primarys oplog
and apply the operations to their data sets.
Although clients cannot write data to secondaries, clients can read data from secondary members. See Read Preference
(page 406) for more information on how clients direct read operations to replica sets.
A secondary can become a primary. If the current primary becomes unavailable, the replica set holds an election to
choose which of the secondaries becomes the new primary.
In the following three-member replica set, the primary becomes unavailable. This triggers an election where one of
the remaining secondaries becomes the new primary.
See Replica Set Elections (page 397) for more details.
1 While replica sets are the recommended solution for production, a replica set can support only 12 members in total. If your deployment requires
more than 12 members, youll need to use master-slave (page 413) replication. Master-slave replication lacks the automatic failover capabilities.

382

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Figure 8.5: Diagram of default routing of reads and writes to the primary.

8.2. Replication Concepts

383

MongoDB Documentation, Release 2.4.9

Figure 8.6: Diagram of an election of a new primary. In a three member replica set with two secondaries, the primary
becomes unreachable. The loss of a primary triggers an election where one of the secondaries becomes the new
primary

Figure 8.7: Diagram of a 3 member replica set that consists of a primary and two secondaries.

384

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Figure 8.8: Diagram of an election of a new primary. In a three member replica set with two secondaries, the primary
becomes unreachable. The loss of a primary triggers an election where one of the secondaries becomes the new
primary

8.2. Replication Concepts

385

MongoDB Documentation, Release 2.4.9

You can configure a secondary member for a specific purpose. You can configure a secondary to:
Prevent it from becoming a primary in an election, which allows it to reside in a secondary data center or to
serve as a cold standby. See Priority 0 Replica Set Members (page 386).
Prevent applications from reading from it, which allows it to run applications that require separation from normal
traffic. See Hidden Replica Set Members (page 387).
Keep a running historical snapshot for use in recovery from certain errors, such as unintentionally deleted
databases. See Delayed Replica Set Members (page 388).
Priority 0 Replica Set Members

A priority 0 member is a secondary that cannot become primary. Priority 0 members cannot trigger elections.
Otherwise these members function as normal secondaries. A priority 0 member maintains a copy of the data set,
accepts read operations, and votes in elections. Configure a priority 0 member to prevent secondaries from becoming
primary, which is particularly useful in multi-data center deployments.
In a three-member replica set, in one data center hosts the primary and a secondary. A second data center hosts one
priority 0 member that cannot become primary.

Figure 8.9: Diagram of a 3 member replica set distributed across two data centers. Replica set includes a priority 0
member.

Priority 0 Members as Standbys A priority 0 member can function as a standby. In some replica sets, it might not
be possible to add a new member in a reasonable amount of time. A standby member keeps a current copy of the data
to be able to replace an unavailable member.
In many cases, you need not set standby to priority 0. However, in sets with varied hardware or geographic distribution
(page 396), a priority 0 standby ensures that only qualified members become primary.
A priority 0 standby may also be valuable for some members of a set with different hardware or workload profiles.
In these cases, deploy a member with priority 0 so it cant become primary. Also consider using an hidden member
(page 387) for this purpose.
If your set already has seven voting members, also configure the member as non-voting (page 400).

386

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Priority 0 Members and Failover When configuring a priority 0 member, consider potential failover patterns,
including all possible network partitions. Always ensure that your main data center contains both a quorum of voting
members and contains members that are eligible to be primary.
Configuration To configure a priority 0 member, see Prevent Secondary from Becoming Primary (page 438).
Hidden Replica Set Members

A hidden member maintains a copy of the primarys data set but is invisible to client applications. Hidden members
are good for workloads with different usage patterns from the other members in the replica set. Hidden members are
always priority 0 members (page 386) and cannot become primary. The db.isMaster() method does not display
hidden members. Hidden members, however, do vote in elections (page 397).
In the following five-member replica set, all four secondary members have copies of the primarys data set, but one of
the secondary members is hidden.

Figure 8.10: Diagram of a 5 member replica set with a hidden priority 0 member.

Behavior
Read Operations Clients will not distribute reads with the appropriate read preference (page 406) to hidden members. As a result, these members receive no traffic other than basic replication. Use hidden members for dedicated
tasks such as reporting and backups. Delayed members (page 388) should be hidden.
In a sharded cluster, mongos do not interact with hidden members.
Voting Hidden members do vote in replica set elections. If you stop a hidden member, ensure that the set has an
active majority or the primary will step down.
For the purposes of backups, you can avoid stopping a hidden member with the db.fsyncLock() and
db.fsyncUnlock() operations to flush all writes and lock the mongod instance for the duration of the backup
operation.
Further Reading For more information about backing up MongoDB databases, see MongoDB Backup Methods
(page 136). To configure a hidden member, see Configure a Hidden Replica Set Member (page 439).
8.2. Replication Concepts

387

MongoDB Documentation, Release 2.4.9

Delayed Replica Set Members

Delayed members contain copies of a replica sets data set. However, a delayed members data set reflects an earlier,
or delayed, state of the set. For example, if the current time is 09:52 and a member has a delay of an hour, the delayed
member has no operation more recent than 08:52.
Because delayed members are a rolling backup or a running historical snapshot of the data set, they may help
you recover from various kinds of human error. For example, a delayed member can make it possible to recover from
unsuccessful application upgrades and operator errors including dropped databases and collections.
Considerations
Requirements Delayed members:
Must be priority 0 (page 386) members. Set the priority to 0 to prevent a delayed member from becoming
primary.
Should be hidden (page 387) members. Always prevent applications from seeing and querying delayed members.
do vote in elections for primary.
Behavior Delayed members apply operations from the oplog on a delay. When choosing the amount of delay,
consider that the amount of delay:
must be is equal to or greater than your maintenance windows.
must be smaller than the capacity of the oplog. For more information on oplog size, see Oplog Size (page 411).
Sharding In sharded clusters, delayed members have limited utility when the balancer is enabled. Because delayed
members replicate chunk migrations with a delay, the state of delayed members in a sharded cluster are not useful for
recovering to a previous state of the sharded cluster if any migrations occur during the delay window.
Example In the following 5-member replica set, the primary and all secondaries have copies of the data set. One
member applies operations with a delay of 3600 seconds, or an hour. This delayed member is also hidden and is a
priority 0 member.
Configuration A delayed member has its priority (page 469) equal to 0, hidden (page 468) equal to true,
and its slaveDelay (page 469) equal to the number of seconds of delay:
{
"_id" : <num>,
"host" : <hostname:port>,
"priority" : 0,
"slaveDelay" : <seconds>,
"hidden" : true
}

To configure a delayed member, see Configure a Delayed Replica Set Member (page 441).

388

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Figure 8.11: Diagram of a 5 member replica set with a hidden delayed priority 0 member.
Replica Set Arbiter
An arbiter does not have a copy of data set and cannot become a primary. Replica sets may have arbiters to add a vote
in elections of for primary (page 397). Arbiters allow replica sets to have an uneven number of members, without the
overhead of a member that replicates data.
Important: Do not run an arbiter on systems that also host the primary or the secondary members of the replica set.
Only add an arbiter to sets with even numbers of members. If you add an arbiter to a set with an odd number of
members, the set may suffer from tied elections. To add an arbiter, see Add an Arbiter to Replica Set (page 431).
Example

For example, in the following replica set, an arbiter allows the set to have an odd number of votes for elections:

Figure 8.12: Diagram of a four member replica set plus an arbiter for odd number of votes.

8.2. Replication Concepts

389

MongoDB Documentation, Release 2.4.9

Security

Authentication When running with auth, arbiters exchange credentials with other members of the set to authenticate. MongoDB encrypts the authentication process. The MongoDB authentication exchange is cryptographically
secure.
Arbiters, use keyfiles to authenticate to the replica set.
Communication The only communication between arbiters and other set members are: votes during elections,
heartbeats, and configuration data. These exchanges are not encrypted.
However, if your MongoDB deployment uses SSL, MongoDB will encrypt all communication between replica set
members. See Connect to MongoDB with SSL (page 252) for more information.
As with all MongoDB components, run arbiters on in trusted network environments.

8.2.2 Replica Set Deployment Architectures

The architecture of a replica set affects the sets capacity and capability. This document provides strategies for replica
set deployments and describes common architectures.
The standard replica set deployment for production system is a three-member replica set. These sets provide redundancy and fault tolerance. Avoid complexity when possible, but let your application requirements dictate the
architecture.
Important: If your application connects to more than one replica set, each set should have a distinct name. Some
drivers group replica set connections by replica set name.

Strategies
Determine the Number of Members

Add members in a replica set according to these strategies.

Deploy an Odd Number of Members An odd number of members ensures that the replica set is always able to
elect a primary. If you have an even number of members, add an arbiter to get an odd number. Arbiters do not store
a copy of the data and require fewer resources. As a result, you may run an arbiter on an application server or other
shared process.
Consider Fault Tolerance Fault tolerance for a replica set is the number of members that can become unavailable
and still leave enough members in the set to elect a primary. In other words, it is the difference between the number
of members in the set and the majority needed to elect a primary. Without a primary, a replica set cannot accept write
operations. Fault tolerance is an effect of replica set size, but the relationship is not direct. See the following table:
Number of Members.
3
4
5
6

Majority Required to Elect a New Primary.

2
3
3
4

Fault Tolerance.
1
1
2
2

Adding a member to the replica set does not always increase the fault tolerance. However, in these cases, additional
members can provide support for dedicated functions, such as backups or reporting.

390

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Use Hidden and Delayed Members for Dedicated Functions Add hidden (page 387) or delayed (page 388) members to support dedicated functions, such as backup or reporting.
Load Balance on Read-Heavy Deployments In a deployment with very high read traffic, you can improve read
throughput by distributing reads to secondary members. As your deployment grows, add or move members to alternate
data centers to improve redundancy and availability.
Always ensure that the main facility is able to elect a primary.
Add Capacity Ahead of Demand The existing members of a replica set must have spare capacity to support adding
a new member. Always add new members before the current demand saturates the capacity of the set.
Determine the Distribution of Members

Distribute Members Geographically To protect your data if your main data center fails, keep at least one member
in an alternate data center. Set these members priority (page 469) to 0 to prevent them from becoming primary.
Keep a Majority of Members in One Location When a replica set has members in multiple data centers, network
partitions can prevent communication between data centers. To replicate data, members must be able to communicate
to other members.
In an election, members must see each other to create a majority. To ensure that the replica set members can confirm
a majority and elect a primary, keep a majority of the sets members in one location.
Target Operations with Tags

Use replica set tags (page 450) to ensure that operations replicate to specific data centers. Tags also support targeting
read operations to specific machines.
See also:
Data Center Awareness (page 159) and Operational Segregation in MongoDB Deployments (page 160).
Use Journaling to Protect Against Power Failures

Enable journaling to protect data against service interruptions. Without journaling MongoDB cannot recover data after
unexpected shutdowns, including power failures and unexpected reboots.
All 64-bit versions of MongoDB after version 2.0 have journaling enabled by default.
Deployment Patterns
The following documents describe common replica set deployment patterns. Other patterns are possible and effective depending on the the applications requirements. If needed, combine features of each architecture in your own
deployment:
Three Member Replica Sets (page 392) Three-member replica sets provide the minimum recommended architecture
for a replica set.
Replica Sets with Four or More Members (page 395) Four or more member replica sets provide greater redundancy
and can support greater distribution of read operations and dedicated functionality.

8.2. Replication Concepts

391

MongoDB Documentation, Release 2.4.9

Geographically Distributed Replica Sets (page 396) Geographically distributed sets include members in multiple locations to protect against facility-specific failures, such as power outages.
Three Member Replica Sets

The minimum architecture of a replica set has three members. A three member replica set can have either three
members that hold data, or two members that hold data and an arbiter.
Primary with Two Secondary Members A replica set with three members that store data has:
One primary (page 382).
Two secondary (page 382) members. Both secondaries can become the primary in an election (page 397).

Figure 8.13: Diagram of a 3 member replica set that consists of a primary and two secondaries.
These deployments provide two complete copies of the data set at all times in addition to the primary. These replica
sets provide additional fault tolerance and high availability (page 397). If the primary is unavailable, the replica set
elects a secondary to be primary and continues normal operation. The old primary rejoins the set when available.
Primary with a Secondary and an Arbiter A three member replica set with a two members that store data has:
One primary (page 382).
One secondary (page 382) member. The secondary can become primary in an election (page 397).
One arbiter (page 389). The arbiter only votes in elections.
Since the arbiter does not hold a copy of the data, these deployments provides only one complete copy of the data.
Arbiters require fewer resources, at the expense of more limited redundancy and fault tolerance.
However, a deployment with a primary, secondary, and an arbiter ensures that a replica set remains available if the
primary or the secondary is unavailable. If the primary is unavailable, the replica set will elect the secondary to be
primary.
See also:
Deploy a Replica Set (page 420).

392

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Figure 8.14: Diagram of an election of a new primary. In a three member replica set with two secondaries, the primary
becomes unreachable. The loss of a primary triggers an election where one of the secondaries becomes the new
primary

Figure 8.15: Diagram of a replica set that consists of a primary, a secondary, and an arbiter.

8.2. Replication Concepts

393

MongoDB Documentation, Release 2.4.9

Figure 8.16: Diagram of an election of a new primary. In a three member replica set with a secondary and an arbiter, the
primary becomes unreachable. The loss of a primary triggers an election where the secondary becomes new primary.

394

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Replica Sets with Four or More Members

Although the standard replica set configuration has three members you can deploy larger sets. Add additional members
to a set to increase redundancy or to add capacity for distributing secondary read operations.
When adding members, ensure that:
The set has an odd number of voting members. If you have an even number of voting members, deploy an
arbiter (page ??) so that the set has an odd number.
The following replica set needs an arbiter to have an odd number of voting members.

Figure 8.17: Diagram of a four member replica set plus an arbiter for odd number of votes.
A replica set can have up to 12 members, 2 but only 7 voting members. See non-voting members (page 400) for
more information.
The following 9 member replica set has 7 voting members and 2 non-voting members.

Figure 8.18: Diagram of a 9 member replica set with the maximum of 7 voting members.
2 While replica sets are the recommended solution for production, a replica set can support only 12 members in total. If your deployment requires
more than 12 members, youll need to use master-slave (page 413) replication. Master-slave replication lacks the automatic failover capabilities.

8.2. Replication Concepts

395

MongoDB Documentation, Release 2.4.9

Members that cannot become primary in a failover have priority 0 configuration (page 386).
For instance, some members that have limited resources or networking constraints and should never be able to
become primary. Configure members that should not become primary to have priority 0 (page 386). In following
replica set, the secondary member in the third data center has a priority of 0:

Figure 8.19: Diagram of a 5 member replica set distributed across three data centers. Replica set includes a priority 0
member.
A majority of the sets members should be in your applications main data center.
See also:
Deploy a Replica Set (page 420), Add an Arbiter to Replica Set (page 431), and Add Members to a Replica Set
(page 433).
Geographically Distributed Replica Sets

Adding members to a replica set in multiple data centers adds redundancy and provides fault tolerance if one data
center is unavailable. Members in additional data centers should have a priority of 0 (page 386) to prevent them from
becoming primary.
For example: the architecture of a geographically distributed replica set may be:
One primary in the main data center.
One secondary member in the main data center. This member can become primary at any time.
One priority 0 (page 386) member in a second data center. This member cannot become primary.
In the following replica set, the primary and one secondary are in Data Center 1, while Data Center 2 has a priority 0
(page 386) secondary that cannot become a primary.
If the primary is unavailable, the replica set will elect a new primary from Data Center 1. If the data centers cannot
connect to each other, the member in Data Center 2 will not become the primary.
If Data Center 1 becomes unavailable, you can manually recover the data set from Data Center 2 with minimal
downtime. With sufficient write concern (page 47), there will be no data loss.
To facilitate elections, the main data center should hold a majority of members. Also ensure that the set has an odd
number of members. If adding a member in another data center results in a set with an even number of members,
deploy an arbiter (page ??). For more information on elections, see Replica Set Elections (page 397).

396

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Figure 8.20: Diagram of a 3 member replica set distributed across two data centers. Replica set includes a priority 0
member.
See also:
Deploy a Geographically Redundant Replica Set (page 425).

8.2.3 Replica Set High Availability

Replica sets provide high availability using automatic failover. Failover allows a secondary members to become
primary if primary is unavailable. Failover, in most situations does not require manual intervention.
Replica set members keep the same data set but are otherwise independent. If the primary becomes unavailable, the
replica set holds an election (page 397) to select a new primary. In some situations, the failover process may require a
rollback (page 401). 3
The deployment of a replica set affects the outcome of failover situations. To support effective failover, ensure that one
facility can elect a primary if needed. Choose the facility that hosts the core application systems to host the majority
of the replica set. Place a majority of voting members and all the members that can become primary in this facility.
Otherwise, network partitions could prevent the set from being able to form a majority.
Failover Processes
The replica set recovers from the loss of a primary by holding an election. Consider the following:
Replica Set Elections (page 397) Elections occur when the primary becomes unavailable and the replica set members
autonomously select a new primary.
Rollbacks During Replica Set Failover (page 401) A rollback reverts write operations on a former primary when the
member rejoins the replica set after a failover.
Replica Set Elections

Replica sets use elections to determine which set member will become primary. Elections occur after initiating a
replica set, and also any time the primary becomes unavailable. The primary is the only member in the set that can
3

Replica sets remove rollback data when needed without intervention. Administrators must apply or discard rollback data manually.

8.2. Replication Concepts

397

MongoDB Documentation, Release 2.4.9

accept write operations. If a primary becomes unavailable, elections allow the set to recover normal operations without
manual intervention. Elections are part of the failover process (page 397).
Important: Elections are essential for independent operation of a replica set; however, elections take time to complete. While an election is in process, the replica set has no primary and cannot accept writes. MongoDB avoids
elections unless necessary.
In the following three-member replica set, the primary is unavailable. The remaining secondaries hold an election to
choose a new primary.

Figure 8.21: Diagram of an election of a new primary. In a three member replica set with two secondaries, the primary
becomes unreachable. The loss of a primary triggers an election where one of the secondaries becomes the new
primary

Factors and Conditions that Affect Elections

Heartbeats Replica set members send heartbeats (pings) to each other every two seconds. If a heartbeat does not
return within 10 seconds, the other members mark the delinquent member as inaccessible.
Priority Comparisons The priority (page 469) setting affects elections. Members will prefer to vote for members with the highest priority value.
398

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Members with a priority value of 0 cannot become primary and do not seek election. For details, see Priority 0 Replica
Set Members (page 386).
A replica set does not hold an election as long as the current primary has the highest priority value and is within 10
seconds of the latest oplog entry in the set. If a higher-priority member catches up to within 10 seconds of the latest
oplog entry of the current primary, the set holds an election in order to provide the higher-priority node a chance to
become primary.
Optime The optime is the timestamp of the last operation that a member applied from the oplog. A replica set
member cannot become primary unless it has the highest (i.e. most recent) optime of any visible member in the set.
Connections A replica set member cannot become primary unless it can connect to a majority of the members in the
replica set. For the purposes of elections, a majority refers to the total number of votes, rather than the total number of
members.
If you have a three-member replica set, where every member has one vote, the set can elect a primary as long as two
members can connect to each other. If two members are unavailable, the remaining member remains a secondary
because it cannot connect to a majority of the sets members. If the remaining member is a primary and two members
become unavailable, the primary steps down and becomes and secondary.
Network Partitions Network partitions affect the formation of a majority for an election. If a primary steps down
and neither portion of the replica set has a majority the set will not elect a new primary. The replica set becomes
read-only.
To avoid this situation, place a majority of instances in one data center and a minority of instances in any other data
centers combined.
Election Mechanics
Election Triggering Events Replica sets hold an election any time there is no primary. Specifically, the following:
the initiation of a new replica set.
a secondary loses contact with a primary. Secondaries call for elections when they cannot see a primary.
a primary steps down.
Note: Priority 0 members (page 386), do not trigger elections, even when they cannot connect to the primary.
A primary will step down:
after receiving the replSetStepDown command.
if one of the current secondaries is eligible for election and has a higher priority.
if primary cannot contact a majority of the members of the replica set.
Important: When a primary steps down, it closes all open client connections, so that clients dont attempt to write
data to a secondary. This helps clients maintain an accurate view of the replica set and helps prevent rollbacks.

8.2. Replication Concepts

399

MongoDB Documentation, Release 2.4.9

Participation in Elections Every replica set member has a priority that helps determine its eligibility to become a
primary. In an election, the replica set elects an eligible member with the highest priority (page 469) value as
primary. By default, all members have a priority of 1 and have an equal chance of becoming primary. In the default,
all members also can trigger an election.
You can set the priority (page 469) value to weight the election in favor of a particular member or group of
members. For example, if you have a geographically distributed replica set (page 396), you can adjust priorities so
that only members in a specific data center can become primary.
The first member to receive the majority of votes becomes primary. By default, all members have a single vote, unless
you modify the votes (page 469) setting. Non-voting members (page 442) have votes (page 469) value of 0.
The state of a member also affects its eligibility to vote. Only members in the following states can vote: PRIMARY,
SECONDARY, RECOVERING, ARBITER, and ROLLBACK.
Important: Do not alter the number of votes in a replica set to control the outcome of an election. Instead, modify
the priority (page 469) value.

Vetoes in Elections All members of a replica set can veto an election, including non-voting members (page 400). A
member will veto an election:
If the member seeking an election is not a member of the voters set.
If the member seeking an election is not up-to-date with the most recent operation accessible in the replica set.
If the member seeking an election has a lower priority than another member in the set that is also eligible for
election.
If a priority 0 member (page 386) 4 is the most current member at the time of the election. In this case, another
eligible member of the set will catch up to the state of this secondary member and then attempt to become
primary.
If the current primary has more recent operations (i.e. a higher optime) than the member seeking election,
from the perspective of the voting member.
If the current primary has the same or more recent operations (i.e. a higher or equal optime) than the member
seeking election.
Non-Voting Members Non-voting members hold copies of the replica sets data and can accept read operations from
client applications. Non-voting members do not vote in elections, but can veto (page 400) an election and become
primary.
Because a replica set can have up to 12 members but only up to seven voting members, non-voting members allow a
replica set to have more than seven members.
For instance, the following nine-member replica set has seven voting members and two non-voting members.
A non-voting member has a votes (page 469) setting equal to 0 in its member configuration:
{
"_id" : <num>
"host" : <hostname:port>,
"votes" : 0
}

Important: Do not alter the number of votes to control which members will become primary. Instead, modify the
4

Remember that hidden (page 387) and delayed (page 388) imply priority 0 (page 386) configuration.

400

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Figure 8.22: Diagram of a 9 member replica set with the maximum of 7 voting members.
priority (page 469) option. Only alter the number of votes in exceptional cases. For example, to permit more than
seven members.
When possible, all members should have only one vote. Changing the number of votes can cause ties, deadlocks, and
the wrong members to become primary.
To configure a non-voting member, see Configure Non-Voting Replica Set Member (page 442).
Rollbacks During Replica Set Failover

A rollback reverts write operations on a former primary when the member rejoins its replica set after a failover.
A rollback is necessary only if the primary had accepted write operations that the secondaries had not successfully
replicated before the primary stepped down. When the primary rejoins the set as a secondary, it reverts, or rolls back,
its write operations to maintain database consistency with the other members.
MongoDB attempts to avoid rollbacks, which should be rare. When a rollback does occur, it is often the result of a
network partition. Secondaries that can not keep up with the throughput of operations on the former primary, increase
the size and impact of the rollback.
A rollback does not occur if the write operations replicate to another member of the replica set before the primary
steps down and if that member remains available and accessible to a majority of the replica set.
Collect Rollback Data When a rollback does occur, administrators must decide whether to apply or ignore the
rollback data. MongoDB writes the rollback data to BSON files in the rollback/ folder under the databases
dbpath directory. The names of rollback files have the following form:
<database>.<collection>.<timestamp>.bson

For example:
records.accounts.2011-05-09T18-10-04.0.bson

Administrators must apply rollback data manually after the member completes the rollback and returns to secondary
status. Use bsondump to read the contents of the rollback files. Then use mongorestore to apply the changes to
the new primary.

8.2. Replication Concepts

401

MongoDB Documentation, Release 2.4.9

Avoid Replica Set Rollbacks To prevent rollbacks, use replica acknowledged write concern (page 49) to guarantee
that the write operations propagate to the members of a replica set.
Rollback Limitations A mongod instance will not rollback more than 300 megabytes of data. If your system must
rollback more than 300 megabytes, you must manually intervene to recover the data. If this is the case, the following
line will appear in your mongod log:
[replica set sync] replSet syncThread: 13410 replSet too much data to roll back

In this situation, save the data directly or force the member to perform an initial sync. To force initial sync, sync from
a current member of the set by deleting the content of the dbpath directory for the member that requires a larger
rollback.
See also:
Replica Set High Availability (page 397) and Replica Set Elections (page 397).

8.2.4 Replica Set Read and Write Semantics

From the perspective of a client application, whether a MongoDB instance is running as a single server (i.e. standalone) or a replica set is transparent.
By default, in MongoDB, read operations to a replica set return results from the primary (page 382) and are consistent
with the last write operation.
Users may configure read preference on a per-connection basis to prefer that the read operations return on the secondary members. If clients configure the read preference to permit secondary reads, read operations cannot return
from secondary members that have not replicated more recent updates or operations. When reading from a secondary,
a query may return data that reflects a previous state.
This behavior is sometimes characterized as eventual consistency because the secondary members state will eventually
reflect the primarys state and MongoDB cannot guarantee strict consistency for read operations from secondary
members.
To guarantee consistency for reads from secondary members, you can configure the client and driver to ensure that
write operations succeed on all members before completing successfully. See Write Concern (page 47) for more
information. Additionally, such configuration can help prevent Rollbacks During Replica Set Failover (page 401)
during a failover.
Note: Sharded clusters where the shards are also replica sets provide the same operational semantics with regards to
write and read operations.
Write Concern for Replica Sets (page 402) Write concern is the guarantee an application requires from MongoDB
to consider a write operation successful.
Read Preference (page 406) Applications specify read preference to control how drivers direct read operations to
members of the replica set.
Read Preference Processes (page 408) With replica sets, read operations may have additional semantics and behavior.
Write Concern for Replica Sets
MongoDBs built-in write concern (page 47) confirms the success of write operations to a replica sets primary.
Write concern uses the getLastError command after write operations to return an object with error information
or confirmation that there are no errors.

402

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

From the perspective of a client application, whether a MongoDB instance is running as a single server (i.e. standalone) or a replica set is transparent. However, replica sets offer some configuration options for write and read
operations. 5
Verify Write Operations

The default write concern confirms write operations only on the primary. You can configure write concern to confirm
write operations to additional replica set members as well by issuing the getLastError command with the w
option.
The w option confirms that write operations have replicated to the specified number of replica set members, including
the primary. You can either specify a number or specify majority, which ensures the write propagates to a majority
of set members.
If you specify a w value greater than the number of members that hold a copy of the data (i.e., greater than the number
of non-arbiter members), the operation blocks until those members become available. This can cause the operation to
block forever. To specify a timeout threshold for the getLastError operation, use the wtimeout argument. A
wtimeout value of 0 means that the operation will never time out.
See getLastError Examples for example invocations.
Modify Default Write Concern

You can configure your own default getLastError behavior for a replica set.
Use the
getLastErrorDefaults (page 470) setting in the replica set configuration (page 467). The following sequence of commands creates a configuration that waits for the write operation to complete on a majority of the set
members before returning:
cfg = rs.conf()
cfg.settings = {}
cfg.settings.getLastErrorDefaults = {w: "majority"}
rs.reconfig(cfg)

The getLastErrorDefaults (page 470) setting affects only those getLastError commands that have no
other arguments.
Note: Use of insufficient write concern can lead to rollbacks (page 401) in the case of replica set failover (page 397).
Always ensure that your operations have specified the required write concern for your application.
See also:
Write Concern (page 47) and connections-write-concern
Custom Write Concerns

You can use replica set tags to create custom write concerns using the getLastErrorDefaults (page 470) and
getLastErrorModes (page 470) replica set settings.
Note: Custom write concern modes specify the field name and a number of distinct values for that field. By contrast,
read preferences use the value of fields in the tag document to direct read operations.
In some cases, you may be able to use the same tags for read preferences and write concerns; however, you may need
to create additional tags for write concerns depending on the requirements of your application.
5

Sharded clusters where the shards are also replica sets provide the same configuration options with regards to write and read operations.

8.2. Replication Concepts

403

MongoDB Documentation, Release 2.4.9

Figure 8.23: Write operation to a replica set with write concern level of w:2 or write to the primary and at least one
secondary.

404

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Single Tag Write Concerns

Consider a five member replica set, where each member has one of the following tag sets:
{
{
{
{
{

"use":
"use":
"use":
"use":
"use":

"reporting" }
"backup" }
"application" }
"application" }
"application" }

You could create a custom write concern mode that will ensure that applicable write operations will not return until
members with two different values of the use tag have acknowledged the write operation. Create the mode with the
following sequence of operations in the mongo shell:
cfg = rs.conf()
cfg.settings = { getLastErrorModes: { use2: { "use": 2 } } }
rs.reconfig(cfg)

To use this mode pass the string use2 to the w option of getLastError as follows:
db.runCommand( { getLastError: 1, w: "use2" } )

Specific Custom Write Concerns

If you have a three member replica with the following tag sets:
{ "disk": "ssd" }
{ "disk": "san" }
{ "disk": "spinning" }

You cannot specify a custom getLastErrorModes (page 470) value to ensure that the write propagates to the san
before returning. However, you may implement this write concern policy by creating the following additional tags, so
that the set resembles the following:
{ "disk": "ssd" }
{ "disk": "san", "disk.san": "san" }
{ "disk": "spinning" }

Then, create a custom getLastErrorModes (page 470) value, as follows:

cfg = rs.conf()
cfg.settings = { getLastErrorModes: { san: { "disk.san": 1 } } }
rs.reconfig(cfg)

To use this mode pass the string san to the w option of getLastError as follows:
db.runCommand( { getLastError: 1, w: "san" } )

This operation will not return until a replica set member with the tag disk.san returns.
You may set a custom write concern mode as the default write concern mode using getLastErrorDefaults
(page 470) replica set as in the following setting:
cfg = rs.conf()
cfg.settings.getLastErrorDefaults = { ssd: 1 }
rs.reconfig(cfg)

8.2. Replication Concepts

405

MongoDB Documentation, Release 2.4.9

See also:
Configure Replica Set Tag Sets (page 450) for further information about replica set reconfiguration and tag sets.
Read Preference
Read preference describes how MongoDB clients route read operations to members of a replica set.
By default, an application directs its read operations to the primary member in a replica set. Reading from the primary
guarantees that read operations reflect the latest version of a document. However, by distributing some or all reads to
secondary members of the replica set, you can improve read throughput or reduce latency for an application that does
not require fully up-to-date data.
Important: You must exercise care when specifying read preferences: modes other than primary (page 476) can
and will return stale data because the secondary queries will not include the most recent write operations to the replica
sets primary.
The following are common use cases for using non-primary (page 476) read preference modes:
Running systems operations that do not affect the front-end application.
Issuing reads to secondaries helps distribute load and prevent operations from affecting the main workload of
the primary. This can be a good choice for reporting and analytics workloads, for example.
Note: Read preferences arent relevant to direct connections to a single mongod instance. However, in order
to perform read operations on a direct connection to a secondary member of a replica set, you must set a read
preference, such as secondary.
Providing local reads for geographically distributed applications.
If you have application servers in multiple data centers, you may consider having a geographically distributed
replica set (page 396) and using a non primary read preference or the nearest (page 477). This allows the
client to read from the lowest-latency members, rather than always reading from the primary.
Maintaining availability during a failover.
Use primaryPreferred (page 476) if you want an application to read from the primary under normal
circumstances, but to allow stale reads from secondaries in an emergency. This provides a read-only mode for
your application during a failover.
Note: In general, do not use primary (page 476) and primaryPreferred (page 476) to provide extra capacity.
Sharding (page 479) increases read and write capacity by distributing read and write operations across a group of
machines, and is often a better strategy for adding capacity.
See
Read Preference Processes (page 408) for more information about the internal application of read preferences.

Read Preference Modes

New in version 2.2.

Important: All read preference modes except primary (page 476) may return stale data because secondaries
replicate operations from the primary with some delay. Ensure that your application can tolerate stale data if you
choose to use a non-primary (page 476) mode.

406

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Figure 8.24: Read operations to a replica set. Default read preference routes the read to the primary. Read preference
of nearest routes the read to the nearest member.

MongoDB drivers (page 95) support five read preference modes.

Read Preference
Description
Mode
primary (page 476)
Default mode. All operations read from the current replica set primary.
primaryPreferred In most situations, operations read from the primary but if it is unavailable, operations
(page 476)
read from secondary members.
secondary
All operations read from the secondary members of the replica set.
(page 476)
secondaryPreferred In most situations, operations read from secondary members but if no secondary
(page 477)
members are available, operations read from the primary.
nearest (page 477)
Operations read from member of the replica set with the least network latency,
irrespective of the members type.
The syntax for specifying the read preference mode is specific to the driver and to the idioms of the host language6 .
Read preference modes are also available to clients connecting to a sharded cluster through a mongos. The mongos
instance obeys specified read preferences when connecting to the replica set that provides each shard in the cluster.
In the mongo shell, the readPref() cursor method provides access to read preferences.
If read operations account for a large percentage of your applications traffic, distributing reads to secondary members
can improve read throughput. However, in most cases sharding (page 484) provides better support for larger scale
operations, as clusters can distribute read and write operations across a group of machines.
6 http://api.mongodb.org/

8.2. Replication Concepts

407

MongoDB Documentation, Release 2.4.9

For more information, see read preference background (page 406) and read preference behavior (page 408). See also
the documentation for your driver7 .
Tag Sets

Tag sets allow you to target read operations to specific members of a replica set.
Custom read preferences and write concerns evaluate tags sets in different ways. Read preferences consider the value
of a tag when selecting a member to read from. Write concerns ignore the value of a tag to when selecting a member,
except to consider whether or not the value is unique.
You can specify tag sets with the following read preference modes:
primaryPreferred (page 476)
secondary (page 476)
secondaryPreferred (page 477)
nearest (page 477)
Tags are not compatible with mode primary (page 476) and, in general, only apply when selecting (page 408) a
secondary member of a set for a read operation. However, the nearest (page 477) read mode, when combined with
a tag set, selects the matching member with the lowest network latency. This member may be a primary or secondary.
All interfaces use the same member selection logic (page 408) to choose the member to which to direct read operations,
basing the choice on read preference mode and tag sets.
For information on configuring tag sets, see the Configure Replica Set Tag Sets (page 450) tutorial.
For more information on how read preference modes (page 476) interact with tag sets, see the documentation for each
read preference mode (page 476).
Read Preference Processes
Changed in version 2.2.
MongoDB drivers use the following procedures to direct operations to replica sets and sharded clusters. To determine
how to route their operations, applications periodically update their view of the replica sets state, identifying which
members are up or down, which member is primary, and verifying the latency to each mongod instance.
Member Selection

Clients, by way of their drivers, and mongos instances for sharded clusters, periodically update their view of the
replica sets state.
When you select non-primary (page 476) read preference, the driver will determine which member to target using
the following process:
1. Assembles a list of suitable members, taking into account member type (i.e. secondary, primary, or all members).
2. Excludes members not matching the tag sets, if specified.
3. Determines which suitable member is the closest to the client in absolute terms.
7 http://api.mongodb.org/

408

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

4. Builds a list of members that are within a defined ping distance (in milliseconds) of the absolute nearest
member.
Applications can configure the threshold used in this stage. The default acceptable latency is 15 milliseconds,
which you can override in the drivers with their own secondaryAcceptableLatencyMS option. For
mongos you can use the --localThreshold or localThreshold runtime options to set this value.
5. Selects a member from these hosts at random. The member receives the read operation.
Drivers can then associate the thread or connection with the selected member. This request association (page 409) is
configurable by the application. See your driver (page 95) documentation about request association configuration and
default behavior.
Request Association

Important: Request association is configurable by the application. See your driver (page 95) documentation about
request association configuration and default behavior.
Because secondary members of a replica set may lag behind the current primary by different amounts, reads for
secondary members may reflect data at different points in time. To prevent sequential reads from jumping around in
time, the driver can associate application threads to a specific member of the set after the first read, thereby preventing
reads from other members. The thread will continue to read from the same member until:
The application performs a read with a different read preference,
The thread terminates, or
The client receives a socket exception, as is the case when theres a network error or when the mongod closes
connections during a failover. This triggers a retry (page 409), which may be transparent to the application.
When using request association, if the client detects that the set has elected a new primary, the driver will discard all
associations between threads and members.
Auto-Retry

Connections between MongoDB drivers and mongod instances in a replica set must balance two concerns:
1. The client should attempt to prefer current results, and any connection should read from the same member of
the replica set as much as possible.
2. The client should minimize the amount of time that the database is inaccessible as the result of a connection
issue, networking problem, or failover in a replica set.
As a result, MongoDB drivers and mongos:
Reuse a connection to specific mongod for as long as possible after establishing a connection to that instance.
This connection is pinned to this mongod.
Attempt to reconnect to a new member, obeying existing read preference modes (page 476), if the connection to
mongod is lost.
Reconnections are transparent to the application itself. If the connection permits reads from secondary members, after reconnecting, the application can receive two sequential reads returning from different secondaries.
Depending on the state of the individual secondary members replication, the documents can reflect the state of
your database at different moments.
Return an error only after attempting to connect to three members of the set that match the read preference mode
(page 476) and tag set (page 408). If there are fewer than three members of the set, the client will error after
connecting to all existing members of the set.
8.2. Replication Concepts

409

MongoDB Documentation, Release 2.4.9

After this error, the driver selects a new member using the specified read preference mode. In the absence of a
specified read preference, the driver uses primary (page 476).
After detecting a failover situation,
possible.

the driver attempts to refresh the state of the replica set as quickly as

Read Preference in Sharded Clusters

Changed in version 2.2: Before version 2.2, mongos did not support the read preference mode semantics (page 476).
In most sharded clusters, each shard consists of a replica set. As such, read preferences are also applicable. With
regard to read preference, read operations in a sharded cluster are identical to unsharded replica sets.
Unlike simple replica sets, in sharded clusters, all interactions with the shards pass from the clients to the mongos
instances that are actually connected to the set members. mongos is then responsible for the application of read
preferences, which is transparent to applications.
There are no configuration changes required for full support of read preference modes in sharded environments, as long
as the mongos is at least version 2.2. All mongos maintain their own connection pool to the replica set members.
As a result:
A request without a specified preference has primary (page 476), the default, unless, the mongos reuses an
existing connection that has a different mode set.
To prevent confusion, always explicitly set your read preference mode.
All nearest (page 477) and latency calculations reflect the connection between the mongos and the mongod
instances, not the client and the mongod instances.
This produces the desired result, because all results must pass through the mongos before returning to the
client.

8.2.5 Replication Processes

Members of a replica set replicate data continuously. First, a member uses initial sync to capture the data set. Then the
member continuously records and applies every operation that modifies the data set. Every member records operations
in its oplog (page 410), which is a capped collection.
Replica Set Oplog (page 410) The oplog records all operations that modify the data in the replica set.
Replica Set Data Synchronization (page 412) Secondaries must replicate all changes accepted by the primary. This
process is the basis of replica set operations.
Replica Set Oplog
The oplog (operations log) is a special capped collection that keeps a rolling record of all operations that modify the
data stored in your databases. MongoDB applies database operations on the primary and then records the operations
on the primarys oplog. The secondary members then copy and apply these operations in an asynchronous process.
All replica set members contain a copy of the oplog, allowing them to maintain the current state of the database.
To facilitate replication, all replica set members send heartbeats (pings) to all other members. Any member can import
oplog entries from any other member.
Whether applied once or multiple times to the target dataset, each operation in the oplog produces the same results, i.e.
each operation in the oplog is idempotent. For proper replication operations, entries in the oplog must be idempotent:
8 When a failover occurs, all members of the set close all client connections that produce a socket error in the driver. This behavior prevents or
minimizes rollback.

410

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

initial sync
post-rollback catch-up
sharding chunk migrations
Oplog Size

When you start a replica set member for the first time, MongoDB creates an oplog of a default size. The size depends
on the architectural details of your operating system.
In most cases, the default oplog size is sufficient. For example, if an oplog is 5% of free disk space and fills up in 24
hours of operations, then secondaries can stop copying entries from the oplog for up to 24 hours without becoming
stale. However, most replica sets have much lower operation volumes, and their oplogs can hold much higher numbers
of operations.
Before mongod creates an oplog, you can specify its size with the oplogSize option. However, after you have
started a replica set member for the first time, you can only change the size of the oplog using the Change the Size of
the Oplog (page 445) procedure.
By default, the size of the oplog is as follows:
For 64-bit Linux, Solaris, FreeBSD, and Windows systems, MongoDB allocates 5% of the available free disk
space to the oplog. If this amount is smaller than a gigabyte, then MongoDB allocates 1 gigabyte of space.
For 64-bit OS X systems, MongoDB allocates 183 megabytes of space to the oplog.
For 32-bit systems, MongoDB allocates about 48 megabytes of space to the oplog.
Workloads that Might Require a Larger Oplog Size

If you can predict your replica sets workload to resemble one of the following patterns, then you might want to create
an oplog that is larger than the default. Conversely, if your application predominantly performs reads and writes only
a small amount of data, you will oplog may be sufficient.
The following workloads might require a larger oplog size.
Updates to Multiple Documents at Once The oplog must translate multi-updates into individual operations in order
to maintain idempotency. This can use a great deal of oplog space without a corresponding increase in data size or
disk use.
Deletions Equal the Same Amount of Data as Inserts If you delete roughly the same amount of data as you insert,
the database will not grow significantly in disk use, but the size of the operation log can be quite large.
Significant Number of In-Place Updates If a significant portion of the workload is in-place updates, the database
records a large number of operations but does not change the quantity of data on disk.
Oplog Status

To view oplog status, including the size and the time range of operations, issue the
db.printReplicationInfo() method. For more information on oplog status, see Check the Size of the
Oplog (page 464).

8.2. Replication Concepts

411

MongoDB Documentation, Release 2.4.9

Under various exceptional situations, updates to a secondarys oplog might lag behind the desired performance time.
Use db.getReplicationInfo() from a secondary member and the replication status output to assess
the current state of replication and determine if there is any unintended replication delay.
See Replication Lag (page 461) for more information.
Replica Set Data Synchronization
In order to maintain up-to-date copies of the shared data set, members of a replica set sync or replicate data from other
members. MongoDB uses two forms of data synchronization: initial sync (page 412) to populate new members with
the full data set, and replication to apply ongoing changes to the entire data set.
Initial Sync

Initial sync copies all the data from one member of the replica set to another member. A member uses initial sync
when the member has no data, such as when the member is new, or when the member has data but is missing a history
of the sets replication.
When you perform an initial sync, MongoDB does the following:
1. Clones all databases. To clone, the mongod queries every collection in each source database and inserts all data
into its own copies of these collections.
2. Applies all changes to the data set. Using the oplog from the source, the mongod updates its data set to reflect
the current state of the replica set.
3. Builds all indexes on all collections.
When the mongod finishes building all index builds, the member can transition to a normal state, i.e. secondary.
To perform an initial sync, see Resync a Member of a Replica Set (page 449).
Replication

Replica set members replicate data continuously after the initial sync. This process keeps the members up to date
with all changes to the replica sets data. In most cases, secondaries synchronize from the primary. Secondaries
may automatically change their sync targets if needed based on changes in the ping time and state of other members
replication.
For a member to sync from another, the buildIndexes (page 468) setting for both members must have the same
value/ buildIndexes (page 468) must be either true or false for both members.
Beginning in version 2.2, secondaries avoid syncing from delayed members (page 388) and hidden members
(page 387).
Validity and Durability

In a replica set, only the primary can accept write operations. Writing only to the primary provides strict consistency
among members.
Journaling provides single-instance write durability. Without journaling, if a MongoDB instance terminates ungracefully, you must assume that the database is in an invalid state.

412

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Multithreaded Replication

MongoDB applies write operations in batches using multiple threads to improve concurrency. MongoDB groups
batches by namespace and applies operations using a group of threads, but always applies the write operations to a
namespace in order.
While applying a batch, MongoDB blocks all reads. As a result, secondaries can never return data that reflects a state
that never existed on the primary.
Pre-Fetching Indexes to Improve Replication Throughput

To help improve the performance of applying oplog entries, MongoDB fetches memory pages that hold affected data
and indexes. This pre-fetch stage minimizes the amount of time MongoDB holds the write lock while applying oplog
entries. By default, secondaries will pre-fetch all Indexes (page 313).
Optionally, you can disable all pre-fetching or only pre-fetch the index on the _id field.
replIndexPrefetch setting for more information.

See the

8.2.6 Master Slave Replication

Important: Replica sets (page 381) replace master-slave replication for most use cases. If possible, use replica
sets rather than master-slave replication for all new production deployments. This documentation remains to support
legacy deployments and for archival purposes only.
In addition to providing all the functionality of master-slave deployments, replica sets are also more robust for production use. Master-slave replication preceded replica sets and made it possible have a large number of non-master
(i.e. slave) nodes, as well as to restrict replicated operations to only a single database; however, master-slave replication provides less redundancy and does not automate failover. See Deploy Master-Slave Equivalent using Replica
Sets (page 415) for a replica set configuration that is equivalent to master-slave replication. If you wish to convert an
existing master-slave deployment to a replica set, see Convert a Master-Slave Deployment to a Replica Set (page 416).
Fundamental Operations
Initial Deployment

To configure a master-slave deployment, start two mongod instances: one in master mode, and the other in slave
mode.
To start a mongod instance in master mode, invoke mongod as follows:
mongod --master --dbpath /data/masterdb/

With the --master option, the mongod will create a local.oplog.$main (page 474) collection, which the operation log that queues operations that the slaves will apply to replicate operations from the master. The --dbpath
is optional.
To start a mongod instance in slave mode, invoke mongod as follows:
mongod --slave --source <masterhostname><:<port>> --dbpath /data/slavedb/

Specify the hostname and port of the master instance to the --source argument. The --dbpath is optional.
For slave instances, MongoDB stores data about the source server in the local.sources (page 474) collection.

8.2. Replication Concepts

413

MongoDB Documentation, Release 2.4.9

Configuration Options for Master-Slave Deployments

As an alternative to specifying the --source run-time option, can add a document to local.sources (page 474)
specifying the master instance, as in the following operation in the mongo shell:
1
2
3

use local
db.sources.find()
db.sources.insert( { host: <masterhostname> <,only: databasename> } );

In line 1, you switch context to the local database. In line 2, the find() operation should return no documents, to
ensure that there are no documents in the sources collection. Finally, line 3 uses db.collection.insert()
to insert the source document into the local.sources (page 474) collection. The model of the local.sources
(page 474) document is as follows:
host
The host field specifies the mastermongod instance, and holds a resolvable hostname, i.e. IP address, or a
name from a host file, or preferably a fully qualified domain name.
You can append <:port> to the host name if the mongod is not running on the default 27017 port.
only
Optional. Specify a name of a database. When specified, MongoDB will only replicate the indicated database.
Operational Considerations for Replication with Master Slave Deployments

Master instances store operations in an oplog which is a capped collection (page 161). As a result, if a slave falls too
far behind the state of the master, it cannot catchup and must re-sync from scratch. Slave may become out of sync
with a master if:
The slave falls far behind the data updates available from that master.
The slave stops (i.e. shuts down) and restarts later after the master has overwritten the relevant operations from
the master.
When slaves, are out of sync, replication stops. Administrators must intervene manually to restart replication. Use the
resync command. Alternatively, the --autoresync allows a slave to restart replication automatically, after ten
second pause, when the slave falls out of sync with the master. With --autoresync specified, the slave will only
attempt to re-sync once in a ten minute period.
To prevent these situations you should specify a larger oplog when you start the master instance, by adding the
--oplogSize option when starting mongod. If you do not specify --oplogSize, mongod will allocate 5%
of available disk space on start up to the oplog, with a minimum of 1GB for 64bit machines and 50MB for 32bit
machines.
Run time Master-Slave Configuration
MongoDB provides a number of run time configuration options for mongod instances in master-slave deployments.
You can specify these options in configuration files (page 146) or on the command-line. See documentation of the
following:
For master nodes:
master
slave
For slave nodes:
source
414

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

only
slaveDelay
Also consider the Master-Slave Replication Command Line Options for related options.
Diagnostics

On a master instance, issue the following operation in the mongo shell to return replication status from the perspective
of the master:
db.printReplicationInfo()

On a slave instance, use the following operation in the mongo shell to return the replication status from the perspective
of the slave:
db.printSlaveReplicationInfo()

Use the serverStatus as in the following operation, to return status of the replication:
db.serverStatus()

See server status repl fields for documentation of the relevant section of output.
Security
When running with auth enabled, in master-slave deployments configure a keyFile so that slave mongod instances can authenticate and communicate with the master mongod instance.
To enable authentication and configure the keyFile add the following option to your configuration file:
keyFile = /srv/mongodb/keyfile

Note: You may chose to set these run-time configuration options using the --keyFile option on the command line.
Setting keyFile enables authentication and specifies a key file for the mongod instances to use when authenticating
to each other. The content of the key file is arbitrary but must be the same on all members of the deployment can
connect to each other.
The key file must be less one kilobyte in size and may only contain characters in the base64 set. The key file must not
have group or world permissions on UNIX systems. Use the following command to use the OpenSSL package to
generate random content for use in a key file:
openssl rand -base64 741

See also:
Security (page 237) for more information about security in MongoDB
Ongoing Administration and Operation of Master-Slave Deployments
Deploy Master-Slave Equivalent using Replica Sets

If you want a replication configuration that resembles master-slave replication, using replica sets replica sets, consider the following replica configuration document. In this deployment hosts <master> and <slave> 9 provide
9

In replica set configurations, the host (page 468) field must hold a resolvable hostname.

8.2. Replication Concepts

415

MongoDB Documentation, Release 2.4.9

replication that is roughly equivalent to a two-instance master-slave deployment:

{
_id : 'setName',
members : [
{ _id : 0, host : "<master>", priority : 1 },
{ _id : 1, host : "<slave>", priority : 0, votes : 0 }
]
}

See Replica Set Configuration (page 467) for more information about replica set configurations.
Convert a Master-Slave Deployment to a Replica Set

To convert a master-slave deployment to a replica set, restart the current master as a one-member replica set. Then
remove the data directors from previous secondaries and add them as new secondaries to the new replica set.
1. To confirm that the current instance is master, run:
db.isMaster()

This should return a document that resembles the following:

{
"ismaster" : true,
"maxBsonObjectSize" : 16777216,
"maxMessageSizeBytes" : 48000000,
"localTime" : ISODate("2013-07-08T20:15:13.664Z"),
"ok" : 1
}

2. Shut down the mongod processes on the master and all slave(s), using the following command while connected
to each instance:
db.adminCommand({shutdown : 1, force : true})

3. Back up your /data/db directories, in case you need to revert to the master-slave deployment.
4. Start the former master with the --replSet option, as in the following:
mongod --replSet <setname>

5. Connect to the mongod with the mongo shell, and initiate the replica set with the following command:
rs.initiate()

When the command returns, you will have successfully deployed a one-member replica set. You can check the
status of your replica set at any time by running the following command:
rs.status()

You can now follow the convert a standalone to a replica set (page 432) tutorial to deploy your replica set, picking up
from the Expand the Replica Set (page 432) section.
Failing over to a Slave (Promotion)

To permanently failover from a unavailable or damaged master (A in the following example) to a slave (B):
1. Shut down A.
416

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

2. Stop mongod on B.
3. Back up and move all data files that begin with local on B from the dbpath.
Warning:
caution.

Removing local.* is irrevocable and cannot be undone. Perform this step with extreme

4. Restart mongod on B with the --master option.

Note: This is a one time operation, and is not reversible. A cannot become a slave of B until it completes a full resync.

Inverting Master and Slave

If you have a master (A) and a slave (B) and you would like to reverse their roles, follow this procedure. The procedure
assumes A is healthy, up-to-date and available.
If A is not healthy but the hardware is okay (power outage, server crash, etc.), skip steps 1 and 2 and in step 8 replace
all of As files with Bs files in step 8.
If A is not healthy and the hardware is not okay, replace A with a new machine. Also follow the instructions in the
previous paragraph.
To invert the master and slave in a deployment:
1. Halt writes on A using the fsync command.
2. Make sure B is up to date with the state of A.
3. Shut down B.
4. Back up and move all data files that begin with local on B from the dbpath to remove the existing
local.sources data.
Warning:
caution.

Removing local.* is irrevocable and cannot be undone. Perform this step with extreme

5. Start B with the --master option.

6. Do a write on B, which primes the oplog to provide a new sync start point.
7. Shut down B. B will now have a new set of data files that start with local.
8. Shut down A and replace all files in the dbpath of A that start with local with a copy of the files in the
dbpath of B that begin with local.
Considering compressing the local files from B while you copy them, as they may be quite large.
9. Start B with the --master option.
10. Start A with all the usual slave options, but include fastsync.
Creating a Slave from an Existing Masters Disk Image

If you can stop write operations to the master for an indefinite period, you can copy the data files from the master to
the new slave and then start the slave with --fastsync.
Warning: Be careful with --fastsync. If the data on both instances is not identical, a discrepancy will exist
forever.

8.2. Replication Concepts

417

MongoDB Documentation, Release 2.4.9

fastsync is a way to start a slave by starting with an existing master disk image/backup. This option declares that
the administrator guarantees the image is correct and completely up-to-date with that of the master. If you have a full
and complete copy of data from a master you can use this option to avoid a full synchronization upon starting the
slave.
Creating a Slave from an Existing Slaves Disk Image

You can just copy the other slaves data file snapshot without any special options. Only take data snapshots when a
mongod process is down or locked using db.fsyncLock().
Resyncing a Slave that is too Stale to Recover

Slaves asynchronously apply write operations from the master that the slaves poll from the masters oplog. The oplog
is finite in length, and if a slave is too far behind, a full resync will be necessary. To resync the slave, connect to a
slave using the mongo and issue the resync command:
use admin
db.runCommand( { resync: 1 } )

This forces a full resync of all data (which will be very slow on a large database). You can achieve the same effect by
stopping mongod on the slave, deleting the entire content of the dbpath on the slave, and restarting the mongod.
Slave Chaining

Slaves cannot be chained. They must all connect to the master directly.
If a slave attempts slave from another slave you will see the following line in the mongod long of the shell:
assertion 13051 tailable cursor requested on non capped collection ns:local.oplog.$main

Correcting a Slaves Source

To change a slaves source, manually modify the slaves local.sources (page 474) collection.
Example
Consider the following: If you accidentally set an incorrect hostname for the slaves source, as in the following
example:
mongod --slave --source prod.mississippi

You can correct this, by restarting the slave without the --slave and --source arguments:
mongod

Connect to this mongod instance using the mongo shell and update the local.sources (page 474) collection,
with the following operation sequence:
use local
db.sources.update( { host : "prod.mississippi" },
{ $set : { host : "prod.mississippi.example.net" } } )

418

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Restart the slave with the correct command line arguments or with no --source option. After configuring
local.sources (page 474) the first time, the --source will have no subsequent effect. Therefore, both of
the following invocations are correct:
mongod --slave --source prod.mississippi.example.net

or
mongod --slave

The slave now polls data from the correct master.

8.3 Replica Set Tutorials

The administration of replica sets includes the initial deployment of the set, adding and removing members to a set,
and configuring the operational parameters and properties of the set. Administrators generally need not intervene in
failover or replication processes as MongoDB automates these functions. In the exceptional situations that require
manual interventions, the tutorials in these sections describe processes such as resyncing a member. The tutorials in
this section form the basis for all replica set administration.
Replica Set Deployment Tutorials (page 420) Instructions for deploying replica sets, as well as adding and removing
members from an existing replica set.
Deploy a Replica Set (page 420) Configure a three-member replica set for either a production system.
Convert a Standalone to a Replica Set (page 432) Convert an existing standalone mongod instance into a
three-member replica set.
Add Members to a Replica Set (page 433) Add a new member to an existing replica set.
Remove Members from Replica Set (page 435) Remove a member from a replica set.
Member Configuration Tutorials (page 437) Tutorials that describe the process for configuring replica set members.
Adjust Priority for Replica Set Member (page 437) Change the precedence given to a replica set members in
an election for primary.
Prevent Secondary from Becoming Primary (page 438) Make a secondary member ineligible for election as
primary.
Configure a Hidden Replica Set Member (page 439) Configure a secondary member to be invisible to applications in order to support significantly different usage, such as a dedicated backups.
Replica Set Maintenance Tutorials (page 444) Procedures and tasks for common operations on active replica set
deployments.
Change the Size of the Oplog (page 445) Increase the size of the oplog which logs operations. In most cases,
the default oplog size is sufficient.
Resync a Member of a Replica Set (page 449) Sync the data on a member. Either perform initial sync on a
new member or resync the data on an existing member that has fallen too far behind to catch up by way of
normal replication.
Change the Size of the Oplog (page 445) Increase the size of the oplog which logs operations. In most cases,
the default oplog size is sufficient.
Force a Member to Become Primary (page 447) Force a replica set member to become primary.
Change Hostnames in a Replica Set (page 457) Update the replica set configuration to reflect changes in
members hostnames.

8.3. Replica Set Tutorials

419

MongoDB Documentation, Release 2.4.9

Troubleshoot Replica Sets (page 461) Describes common issues and operational challenges for replica sets. For additional diagnostic information, see FAQ: MongoDB Diagnostics (page 587).

8.3.1 Replica Set Deployment Tutorials

The following tutorials provide information in deploying replica sets.
Deploy a Replica Set (page 420) Configure a three-member replica set for either a production system.
Deploy a Replica Set for Testing and Development (page 423) Configure a three-member replica set for either a development and testing systems.
Deploy a Geographically Redundant Replica Set (page 425) Create a geographically redundant replica set to protect
against location-centered availability limitations (e.g. network and power interruptions).
Add an Arbiter to Replica Set (page 431) Add an arbiter give a replica set an odd number of voting members to
prevent election ties.
Convert a Standalone to a Replica Set (page 432) Convert an existing standalone mongod instance into a threemember replica set.
Add Members to a Replica Set (page 433) Add a new member to an existing replica set.
Remove Members from Replica Set (page 435) Remove a member from a replica set.
Replace a Replica Set Member (page 437) Update the replica set configuration when the hostname of a members
corresponding mongod instance has changed.
Deploy a Replica Set
This tutorial describes how to create a three-member replica set from three existing mongod instances.
If you wish to deploy a replica set from a single MongoDB instance, see Convert a Standalone to a Replica Set
(page 432). For more information on replica set deployments, see the Replication (page 377) and Replica Set Deployment Architectures (page 390) documentation.
Overview

Three member replica sets provide enough redundancy to survive most network partitions and other system failures.
These sets also have sufficient capacity for many distributed read operations. Replica sets should always have an odd
number of members. This ensures that elections (page 397) will proceed smoothly. For more about designing replica
sets, see the Replication overview (page 377).
The basic procedure is to start the mongod instances that will become members of the replica set, configure the replica
set itself, and then add the mongod instances to it.
Requirements

For production deployments, you should maintain as much separation between members as possible by hosting the
mongod instances on separate machines. When using virtual machines for production deployments, you should place
each mongod instance on a separate host server serviced by redundant power circuits and redundant network paths.
Before you can deploy a replica set, you must install MongoDB on each system that will be part of your replica set. If
you have not already installed MongoDB, see the installation tutorials (page 3).

420

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Before creating your replica set, you should verify that your network configuration allows all possible connections
between each member. For a successful replica set deployment, every member must be able to connect to every other
member. For instructions on how to check your connection, see Test Connections Between all Members (page 463).
Procedure

Each member of the replica set resides on its own machine and all of the MongoDB processes bind to port
27017 (the standard MongoDB port).
Each member of the replica set must be accessible by way of resolvable DNS or hostnames, as in the following
scheme:
mongodb0.example.net
mongodb1.example.net
mongodb2.example.net
mongodbn.example.net
You will need to either configure your DNS names appropriately, or set up your systems /etc/hosts file to
reflect this configuration.
Ensure that network traffic can pass between all members in the network securely and efficiently. Consider the
following:
Establish a virtual private network. Ensure that your network topology routes all traffic between members
within a single site over the local area network.
Configure authentication using auth and keyFile, so that only servers and processes with authentication can connect to the replica set.
Configure networking and firewall rules so that only traffic (incoming and outgoing packets) on the default
MongoDB port (e.g. 27017) from within your deployment is permitted.
For more information on security and firewalls, see Inter-Process Authentication (page 240).
You must specify the run time configuration on each system in a configuration file stored in
/etc/mongodb.conf or a related location. Do not specify the sets configuration in the mongo shell.
Use the following configuration for each of your MongoDB instances. You should set values that are appropriate
for your systems, as needed:
port = 27017
bind_ip = 10.8.0.10
dbpath = /srv/mongodb/
fork = true
replSet = rs0

The dbpath indicates where you want mongod to store data files. The dbpath must exist before you start
mongod. If it does not exist, create the directory and ensure mongod has permission to read and write data to
this path. For more information on permissions, see the security operations documentation (page 238).
Modifying bind_ip ensures that mongod will only listen for connections from applications on the configured
address.
For more information about the run time options used above and other configuration options, see
http://docs.mongodb.org/manualreference/configuration-options.
8.3. Replica Set Tutorials

421

MongoDB Documentation, Release 2.4.9

To deploy a production replica set:

1. Start a mongod instance on each system that will be part of your replica set. Specify the same replica set name
on each instance. For additional mongod configuration options specific to replica sets, see cli-mongod-replicaset.
Important: If your application connects to more than one replica set, each set should have a distinct name.
Some drivers group replica set connections by replica set name.
If you use a configuration file, then start each mongod instance with a command that resembles the following:
mongod --config /etc/mongodb.conf

Change /etc/mongodb.conf to the location of your configuration file.

Note: You will likely want to use and configure a control script to manage this process in production deployments. Control scripts are beyond the scope of this document.
2. Open a mongo shell connected to one of the hosts by issuing the following command:
mongo

3. Use rs.initiate() to initiate a replica set consisting of the current member and using the default configuration, as follows:
rs.initiate()

4. Display the current replica configuration (page 467):

rs.conf()

The replica set configuration object resembles the following

{
"_id" : "rs0",
"version" : 4,
"members" : [
{
"_id" : 1,
"host" : "mongodb0.example.net:27017"
}
]
}

1. In the mongo shell connected to the primary, add the remaining members to the replica set using rs.add()
in the mongo shell on the current primary (in this example, mongodb0.example.net). The commands
should resemble the following:
rs.add("mongodb1.example.net")
rs.add("mongodb2.example.net")

When complete, you should have a fully functional replica set. The new replica set will elect a primary.
Check the status of your replica set at any time with the rs.status() operation.
See also:
The documentation of the following shell functions for more information:
rs.initiate()

422

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

rs.conf()
rs.reconfig()
rs.add()
Refer to Replica Set Read and Write Semantics (page 402) for a detailed explanation of read and write semantics in
MongoDB.
Deploy a Replica Set for Testing and Development
Note: This tutorial provides instructions for deploying a replica set in a development or test environment. For a
production deployment, refer to the Deploy a Replica Set (page 420) tutorial.
This tutorial describes how to create a three-member replica set from three existing mongod instances.
If you wish to deploy a replica set from a single MongoDB instance, see Convert a Standalone to a Replica Set
(page 432). For more information on replica set deployments, see the Replication (page 377) and Replica Set Deployment Architectures (page 390) documentation.
Overview

Three member replica sets provide enough redundancy to survive most network partitions and other system failures.
These sets also have sufficient capacity for many distributed read operations. Replica sets should always have an odd
number of members. This ensures that elections (page 397) will proceed smoothly. For more about designing replica
sets, see the Replication overview (page 377).
The basic procedure is to start the mongod instances that will become members of the replica set, configure the replica
set itself, and then add the mongod instances to it.
Requirements

For test and development systems, you can run your mongod instances on a local system, or within a virtual instance.
Before you can deploy a replica set, you must install MongoDB on each system that will be part of your replica set. If
you have not already installed MongoDB, see the installation tutorials (page 3).
Before creating your replica set, you should verify that your network configuration allows all possible connections
between each member. For a successful replica set deployment, every member must be able to connect to every other
member. For instructions on how to check your connection, see Test Connections Between all Members (page 463).
Procedure

Important: These instructions should only be used for test or development deployments.
The examples in this procedure create a new replica set named rs0.
Important: If your application connects to more than one replica set, each set should have a distinct
name. Some drivers group replica set connections by replica set name.
You will begin by starting three mongod instances as members of a replica set named rs0.
1. Create the necessary data directories for each member by issuing a command similar to the following:

8.3. Replica Set Tutorials

423

MongoDB Documentation, Release 2.4.9

mkdir -p /srv/mongodb/rs0-0 /srv/mongodb/rs0-1 /srv/mongodb/rs0-2

This will create directories called rs0-0, rs0-1, and rs0-2, which will contain the instances database files.
2. Start your mongod instances in their own shell windows by issuing the following commands:
First member:
mongod --port 27017 --dbpath /srv/mongodb/rs0-0 --replSet rs0 --smallfiles --oplogSize 128

Second member:
mongod --port 27018 --dbpath /srv/mongodb/rs0-1 --replSet rs0 --smallfiles --oplogSize 128

Third member:
mongod --port 27019 --dbpath /srv/mongodb/rs0-2 --replSet rs0 --smallfiles --oplogSize 128

This starts each instance as a member of a replica set named rs0, each running on a distinct port, and specifies
the path to your data directory with the --dbpath setting. If you are already using the suggested ports, select
different ports.
The --smallfiles and --oplogSize settings reduce the disk space that each mongod
instance uses.
This is ideal for testing and development deployments as it prevents overloading your machine.
For more information on these and other configuration options, see
http://docs.mongodb.org/manualreference/configuration-options.
3. Connect to one of your mongod instances through the mongo shell. You will need to indicate which instance
by specifying its port number. For the sake of simplicity and clarity, you may want to choose the first one, as in
the following command;
mongo --port 27017

4. In the mongo shell, use rs.initiate() to initiate the replica set. You can create a replica set configuration
object in the mongo shell environment, as in the following example:
rsconf = {
_id: "rs0",
members: [
{
_id: 0,
host: "<hostname>:27017"
}
]
}

replacing <hostname> with your systems hostname, and then pass the rsconf file to rs.initiate() as
follows:
rs.initiate( rsconf )

5. Display the current replica configuration (page 467) by issuing the following command:
rs.conf()

The replica set configuration object resembles the following

{
"_id" : "rs0",
"version" : 4,
"members" : [

424

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

{
"_id" : 1,
"host" : "localhost:27017"
}
]
}

6. In the mongo shell connected to the primary, add the second and third mongod instances to the replica set
using the rs.add() method. Replace <hostname> with your systems hostname in the following examples:
rs.add("<hostname>:27018")
rs.add("<hostname>:27019")

When complete, you should have a fully functional replica set. The new replica set will elect a primary.
Check the status of your replica set at any time with the rs.status() operation.
See also:
The documentation of the following shell functions for more information:
rs.initiate()
rs.conf()
rs.reconfig()
rs.add()
You may also consider the simple setup script10 as an example of a basic automatically-configured replica set.
Refer to Replica Set Read and Write Semantics (page 402) for a detailed explanation of read and write semantics in
MongoDB.
Deploy a Geographically Redundant Replica Set
This tutorial outlines the process for deploying a replica set with members in multiple locations. The tutorial addresses
three-member sets, four-member sets, and sets with more than four members.
For appropriate background, see Replication (page 377) and Replica Set Deployment Architectures (page 390). For
related tutorials, see Deploy a Replica Set (page 420) and Add Members to a Replica Set (page 433).
Overview

While replica sets provide basic protection against single-instance failure, replica sets whose members are all located
in a single facility are susceptible to errors in that facility. Power outages, network interruptions, and natural disasters
are all issues that can affect replica sets whose members are colocated. To protect against these classes of failures,
deploy a replica set with one or more members in a geographically distinct facility or data center to provide redundancy.
Requirements

In general, the requirements for any geographically redundant replica set are as follows:
Ensure that a majority of the voting members (page 400) are within a primary facility, Site A. This includes
priority 0 members (page 386) and arbiters (page 389). Deploy other members in secondary facilities, Site B,
Site C, etc., to provide additional copies of the data. See Determine the Distribution of Members (page 391)
for more information on the voting requirements for geographically redundant replica sets.
10 https://github.com/mongodb/mongo-snippets/blob/master/replication/simple-setup.py

8.3. Replica Set Tutorials

425

MongoDB Documentation, Release 2.4.9

If you deploy a replica set with an even number of members, deploy an arbiter (page 389) on Site A. The arbiter
must be on site A to keep the majority there.
For instance, for a three-member replica set you need two instances in a Site A, and one member in a secondary facility,
Site B. Site A should be the same facility or very close to your primary application infrastructure (i.e. application
servers, caching layer, users, etc.)
A four-member replica set should have at least two members in Site A, with the remaining members in one or more
secondary sites, as well as a single arbiter in Site A.
For all configurations in this tutorial, deploy each replica set member on a separate system. Although you may deploy
more than one replica set member on a single system, doing so reduces the redundancy and capacity of the replica set.
Such deployments are typically for testing purposes and beyond the scope of this tutorial.
This tutorial assumes you have installed MongoDB on each system that will be part of your replica set. If you have
not already installed MongoDB, see the installation tutorials (page 3).
Procedures

General Considerations
Each member of the replica set resides on its own machine and all of the MongoDB processes bind to port
27017 (the standard MongoDB port).
Each member of the replica set must be accessible by way of resolvable DNS or hostnames, as in the following
scheme:
mongodb0.example.net
mongodb1.example.net
mongodb2.example.net
mongodbn.example.net
You will need to either configure your DNS names appropriately, or set up your systems /etc/hosts file to
reflect this configuration.
Ensure that network traffic can pass between all members in the network securely and efficiently. Consider the
following:
Establish a virtual private network. Ensure that your network topology routes all traffic between members
within a single site over the local area network.
Configure authentication using auth and keyFile, so that only servers and processes with authentication can connect to the replica set.
Configure networking and firewall rules so that only traffic (incoming and outgoing packets) on the default
MongoDB port (e.g. 27017) from within your deployment is permitted.
For more information on security and firewalls, see Inter-Process Authentication (page 240).
You must specify the run time configuration on each system in a configuration file stored in
/etc/mongodb.conf or a related location. Do not specify the sets configuration in the mongo shell.
Use the following configuration for each of your MongoDB instances. You should set values that are appropriate
for your systems, as needed:
port = 27017
bind_ip = 10.8.0.10
dbpath = /srv/mongodb/

426

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

fork = true
replSet = rs0

The dbpath indicates where you want mongod to store data files. The dbpath must exist before you start
mongod. If it does not exist, create the directory and ensure mongod has permission to read and write data to
this path. For more information on permissions, see the security operations documentation (page 238).
Modifying bind_ip ensures that mongod will only listen for connections from applications on the configured
address.
For more information about the run time options used above and other configuration options, see
http://docs.mongodb.org/manualreference/configuration-options.

Figure 8.25: Diagram of a 3 member replica set distributed across two data centers. Replica set includes a priority 0
member.
Deploy a Geographically Redundant Three-Member Replica Set
1. Start a mongod instance on each system that will be part of your replica set. Specify the same replica set name
on each instance. For additional mongod configuration options specific to replica sets, see cli-mongod-replicaset.
Important: If your application connects to more than one replica set, each set should have a distinct name.
Some drivers group replica set connections by replica set name.
If you use a configuration file, then start each mongod instance with a command that resembles the following:
mongod --config /etc/mongodb.conf

Change /etc/mongodb.conf to the location of your configuration file.

Note: You will likely want to use and configure a control script to manage this process in production deployments. Control scripts are beyond the scope of this document.
2. Open a mongo shell connected to one of the hosts by issuing the following command:
mongo

8.3. Replica Set Tutorials

427

MongoDB Documentation, Release 2.4.9

3. Use rs.initiate() to initiate a replica set consisting of the current member and using the default configuration, as follows:
rs.initiate()

4. Display the current replica configuration (page 467):

rs.conf()

The replica set configuration object resembles the following

{
"_id" : "rs0",
"version" : 4,
"members" : [
{
"_id" : 1,
"host" : "mongodb0.example.net:27017"
}
]
}

1. In the mongo shell connected to the primary, add the remaining members to the replica set using rs.add()
in the mongo shell on the current primary (in this example, mongodb0.example.net). The commands
should resemble the following:
rs.add("mongodb1.example.net")
rs.add("mongodb2.example.net")

When complete, you should have a fully functional replica set. The new replica set will elect a primary.
6. Make sure that you have configured the member located
mongodb2.example.net) as a priority 0 member (page 386):

in

Site

(in

this

example,

(a) Issue the following command to determine the members (page 467) array position for the member:
rs.conf()

(b) In the members (page 467) array, save the position of the member whose priority you wish to change.
The example in the next step assumes this value is 2, for the third item in the list. You must record array
position, not _id, as these ordinals will be different if you remove a member.
(c) In the mongo shell connected to the replica sets primary, issue a command sequence similar to the following:
cfg = rs.conf()
cfg.members[2].priority = 0
rs.reconfig(cfg)

When the operations return, mongodb2.example.net has a priority of 0. It cannot become primary.
Note: The rs.reconfig() shell method can force the current primary to step down, causing an
election. When the primary steps down, all clients will disconnect. This is the intended behavior. While
most elections complete within a minute, always make sure any replica configuration changes occur during
scheduled maintenance periods.
After these commands return, you have a geographically redundant three-member replica set.
Check the status of your replica set at any time with the rs.status() operation.
See also:

428

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

The documentation of the following shell functions for more information:

rs.initiate()
rs.conf()
rs.reconfig()
rs.add()
Refer to Replica Set Read and Write Semantics (page 402) for a detailed explanation of read and write semantics in
MongoDB.
Deploy a Geographically Redundant Four-Member Replica Set A geographically redundant four-member deployment has two additional considerations:
One host (e.g. mongodb3.example.net) must be an arbiter. This host can run on a system that is also used
for an application server or on the same machine as another MongoDB process.
You must decide how to distribute your systems. There are three possible architectures for the four-member
replica set:
Three members in Site A, one priority 0 member (page 386) in Site B, and an arbiter in Site A.
Two members in Site A, two priority 0 members (page 386) in Site B, and an arbiter in Site A.
Two members in Site A, one priority 0 member in Site B, one priority 0 member in Site C, and an arbiter
in site A.
In most cases, the first architecture is preferable because it is the least complex.
To deploy a geographically redundant four-member set:
1. Start a mongod instance on each system that will be part of your replica set. Specify the same replica set name
on each instance. For additional mongod configuration options specific to replica sets, see cli-mongod-replicaset.
Important: If your application connects to more than one replica set, each set should have a distinct name.
Some drivers group replica set connections by replica set name.
If you use a configuration file, then start each mongod instance with a command that resembles the following:
mongod --config /etc/mongodb.conf

Change /etc/mongodb.conf to the location of your configuration file.

Note: You will likely want to use and configure a control script to manage this process in production deployments. Control scripts are beyond the scope of this document.
2. Open a mongo shell connected to one of the hosts by issuing the following command:
mongo

3. Use rs.initiate() to initiate a replica set consisting of the current member and using the default configuration, as follows:
rs.initiate()

4. Display the current replica configuration (page 467):

8.3. Replica Set Tutorials

429

MongoDB Documentation, Release 2.4.9

rs.conf()

The replica set configuration object resembles the following

{
"_id" : "rs0",
"version" : 4,
"members" : [
{
"_id" : 1,
"host" : "mongodb0.example.net:27017"
}
]
}

5. Add the remaining members to the replica set using rs.add() in a mongo shell connected to the current
primary. The commands should resemble the following:
rs.add("mongodb1.example.net")
rs.add("mongodb2.example.net")
rs.add("mongodb3.example.net")

When complete, you should have a fully functional replica set. The new replica set will elect a primary.
6. In the same shell session, issue the following command to add the arbiter (e.g. mongodb4.example.net):
rs.addArb("mongodb4.example.net")

7. Make sure that you have configured each member located outside of Site A (e.g. mongodb3.example.net)
as a priority 0 member (page 386):
(a) Issue the following command to determine the members (page 467) array position for the member:
rs.conf()

(b) In the members (page 467) array, save the position of the member whose priority you wish to change.
The example in the next step assumes this value is 2, for the third item in the list. You must record array
position, not _id, as these ordinals will be different if you remove a member.
(c) In the mongo shell connected to the replica sets primary, issue a command sequence similar to the following:
cfg = rs.conf()
cfg.members[2].priority = 0
rs.reconfig(cfg)

When the operations return, mongodb2.example.net has a priority of 0. It cannot become primary.
Note: The rs.reconfig() shell method can force the current primary to step down, causing an
election. When the primary steps down, all clients will disconnect. This is the intended behavior. While
most elections complete within a minute, always make sure any replica configuration changes occur during
scheduled maintenance periods.
After these commands return, you have a geographically redundant four-member replica set.
Check the status of your replica set at any time with the rs.status() operation.
See also:
The documentation of the following shell functions for more information:

430

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

rs.initiate()
rs.conf()
rs.reconfig()
rs.add()
Refer to Replica Set Read and Write Semantics (page 402) for a detailed explanation of read and write semantics in
MongoDB.
Deploy a Geographically Redundant Set with More than Four Members The above procedures detail the steps
necessary for deploying a geographically redundant replica set. Larger replica set deployments follow the same steps,
but have additional considerations:
Never deploy more than seven voting members.
If you have an even number of members, use the procedure for a four-member set (page 429)). Ensure that
a single facility, Site A, always has a majority of the members by deploying the arbiter in that site. For
example, if a set has six members, deploy at least three voting members in addition to the arbiter in Site A, and
the remaining members in alternate sites.
If you have an odd number of members, use the procedure for a three-member set (page 427). Ensure that a
single facility, Site A always has a majority of the members of the set. For example, if a set has five members,
deploy three members within Site A and two members in other facilities.
If you have a majority of the members of the set outside of Site A and the network partitions to prevent communication between sites, the current primary in Site A will step down, even if none of the members outside of
Site A are eligible to become primary.
Add an Arbiter to Replica Set
Arbiters are mongod instances that are part of replica set but do not hold data. Arbiters participate in elections
(page 397) in order to break ties. If a replica set has an even number of members, add an arbiter.
Arbiters have minimal resource requirements and do not require dedicated hardware. You can deploy an arbiter on an
application server, monitoring host.
Important: Do not run an arbiter on the same system as a member of the replica set.

Add an Arbiter

1. Create a data directory (e.g. dbpath) for the arbiter. The mongod instance uses the directory for configuration
data. The directory will not hold the data set. For example, create the /data/arb directory:
mkdir /data/arb

2. Start the arbiter. Specify the data directory and the replica set name. The following, starts an arbiter using the
/data/arb dbpath for the rs replica set:
mongod --port 30000 --dbpath /data/arb --replSet rs

3. Connect to the primary and add the arbiter to the replica set. Use the rs.addArb() method, as in the following
example:
rs.addArb("m1.example.net:30000")

This operation adds the arbiter running on port 30000 on the m1.example.net host.
8.3. Replica Set Tutorials

431

MongoDB Documentation, Release 2.4.9

Convert a Standalone to a Replica Set

This tutorial describes the process for converting a standalone mongod instance into a three-member replica set. Use
standalone instances for testing and development, but always use replica sets in production. To install a standalone
instance, see the installation tutorials (page 3).
To deploy a replica set without using a pre-existing mongod instance, see Deploy a Replica Set (page 420).
Procedure

1. Shut down the standalone mongod instance.

2. Restart the instance. Use the --replSet option to specify the name of the new replica set.
For example, the following command starts a standalone instance as a member of a new replica set named rs0.
The command uses the standalones existing database path of /srv/mongodb/db0:
mongod --port 27017 --dbpath /srv/mongodb/db0 --replSet rs0

Important: If your application connects to more than one replica set, each set should have a distinct name.
Some drivers group replica set connections by replica set name.

For more information on configuration options, see http://docs.mongodb.org/manualreference/configuration

and the mongod manual page.
3. Connect to the mongod instance.
4. Use rs.initiate() to initiate the new replica set:
rs.initiate()

The replica set is now operational.

To view the replica set configuration, use rs.conf().
rs.status().

To check the status of the replica set, use

Expand the Replica Set Add additional replica set members by doing the following:
1. On two distinct systems, start two new standalone mongod instances. For information on starting a standalone
instance, see the installation tutorial (page 3) specific to your environment.
2. On your connection to the original mongod instance (the former standalone instance), issue a command in the
following form for each new instance to add to the replica set:
rs.add("<hostname><:port>")

Replace <hostname> and <port> with the resolvable hostname and port of the mongod instance to add to
the set. For more information on adding a host to a replica set, see Add Members to a Replica Set (page 433).
Sharding Considerations If the new replica set is part of a sharded cluster, change the shard host information in
the config database by doing the following:
1. Connect to one of the sharded clusters mongos instances and issue a command in the following form:

db.getSiblingDB("config").shards.save( {_id: "<name>", host: "<replica-set>/<member,><member,><.

Replace <name> with the name of the shard. Replace <replica-set> with the name of the replica set.
Replace <member,><member,><> with the list of the members of the replica set.
432

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

2. Restart all mongos instances. If possible, restart all components of the replica sets (i.e., all mongos and all
shard mongod instances).
Add Members to a Replica Set
Overview

This tutorial explains how to add an additional member to an existing replica set. For background on replication
deployment patterns, see the Replica Set Deployment Architectures (page 390) document.
Maximum Voting Members A replica set can have a maximum of seven voting members (page 397). To add
a member to a replica set that already has seven votes, you must either add the member as a non-voting member
(page 400) or remove a vote from an existing member (page 469).
Control Scripts In production deployments you can configure a control script to manage member processes.
Existing Members You can use these procedures to add new members to an existing set. You can also use the same
procedure to re-add a removed member. If the removed members data is still relatively recent, it can recover and
catch up easily.
Data Files If you have a backup or snapshot of an existing member, you can move the data files (e.g. the dbpath
directory) to a new system and use them to quickly initiate a new member. The files must be:
A valid copy of the data files from a member of the same replica set. See Backup and Restore with Filesystem
Snapshots (page 190) document for more information.
Important: Always use filesystem snapshots to create a copy of a member of the existing replica set. Do not
use mongodump and mongorestore to seed a new replica set member.
More recent than the oldest operation in the primarys oplog. The new member must be able to become current
by applying operations from the primarys oplog.
Requirements

1. An active replica set.

2. A new MongoDB system capable of supporting your data set, accessible by the active replica set through the
network.
Otherwise, use the MongoDB installation tutorial (page 3) and the Deploy a Replica Set (page 420) tutorials.
Procedures

Prepare the Data Directory Before adding a new member to an existing replica set, prepare the new members data
directory using one of the following strategies:
Make sure the new members data directory does not contain data. The new member will copy the data from an
existing member.
If the new member is in a recovering state, it must exit and become a secondary before MongoDB can copy all
data as part of the replication process. This process takes time but does not require administrator intervention.
8.3. Replica Set Tutorials

433

MongoDB Documentation, Release 2.4.9

Manually copy the data directory from an existing member. The new member becomes a secondary member
and will catch up to the current state of the replica set. Copying the data over may shorten the amount of time
for the new member to become current.
Ensure that you can copy the data directory to the new member and begin replication within the window allowed
by the oplog (page 411). Otherwise, the new instance will have to perform an initial sync, which completely
resynchronizes the data, as described in Resync a Member of a Replica Set (page 449).
Use db.printReplicationInfo() to check the current state of replica set members with regards to the
oplog.
For background on replication deployment patterns, see the Replica Set Deployment Architectures (page 390) document.
Add a Member to an Existing Replica Set
1. Start the new mongod instance. Specify the data directory and the replica set name. The following example
specifies the /srv/mongodb/db0 data directory and the rs0 replica set:
mongod --dbpath /srv/mongodb/db0 --replSet rs0

Take note of the host name and port information for the new mongod instance.
For more information on configuration options, see the mongod manual page.
Optional
You can specify the data directory and replica set in the mongo.conf configuration file, and start the
mongod with the following command:
mongod --config /etc/mongodb.conf

2. Connect to the replica sets primary.

You can only add members while connected to the primary. If you do not know which member is the primary,
log into any member of the replica set and issue the db.isMaster() command.
3. Use rs.add() to add the new member to the replica set.
mongodb3.example.net, issue the following command:

For example, to add a member at host

rs.add("mongodb3.example.net")

You can include the port number, depending on your setup:

rs.add("mongodb3.example.net:27017")

4. Verify that the member is now part of the replica set. Call the rs.conf() method, which displays the replica
set configuration (page 467):
rs.conf()

To view replica set status, issue the rs.status() method. For a description of the status fields, see
http://docs.mongodb.org/manualreference/command/replSetGetStatus.
Configure and Add a Member You can add a member to a replica set by passing to the rs.add() method a
members (page 467) document. The document must be in the form of a local.system.replset.members
(page 467) document. These documents define a replica set member in the same form as the replica set configuration
document (page 467).

434

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Important: Specify a value for the _id field of the members (page 467) document. MongoDB does not automatically populate the _id field in this case. Finally, the members (page 467) document must declare the host value.
All other fields are optional.
Example
To add a member with the following configuration:
an _id of 1.
a hostname and port number (page 468) of mongodb3.example.net:27017.
a priority (page 469) value within the replica set of 0.
a configuration as hidden (page 468),
Issue the following:
rs.add({_id: 1, host: "mongodb3.example.net:27017", priority: 0, hidden: true})

Remove Members from Replica Set

To remove a member of a replica set use either of the following procedures.
Remove a Member Using rs.remove()

1. Shut down the mongod instance for the member you wish to remove. To shut down the instance, connect using
the mongo shell and the db.shutdownServer() method.
2. Connect to the replica sets current primary. To determine the current primary, use db.isMaster() while
connected to any member of the replica set.
3. Use rs.remove() in either of the following forms to remove the member:
rs.remove("mongod3.example.net:27017")
rs.remove("mongod3.example.net")

MongoDB disconnects the shell briefly as the replica set elects a new primary. The shell then automatically
reconnects. The shell displays a DBClientCursor::init call() failed error even though the command succeeds.
Remove a Member Using rs.reconfig()

To remove a member you can manually edit the replica set configuration document (page 467), as described here.
1. Shut down the mongod instance for the member you wish to remove. To shut down the instance, connect using
the mongo shell and the db.shutdownServer() method.
2. Connect to the replica sets current primary. To determine the current primary, use db.isMaster() while
connected to any member of the replica set.
3. Issue the rs.conf() method to view the current configuration document and determine the position in the
members array of the member to remove:
Example
mongod_C.example.net is in position 2 of the following configuration file:

8.3. Replica Set Tutorials

435

MongoDB Documentation, Release 2.4.9

{
"_id" : "rs",
"version" : 7,
"members" : [
{
"_id" : 0,
"host" : "mongod_A.example.net:27017"
},
{
"_id" : 1,
"host" : "mongod_B.example.net:27017"
},
{
"_id" : 2,
"host" : "mongod_C.example.net:27017"
}
]
}

4. Assign the current configuration document to the variable cfg:

cfg = rs.conf()

5. Modify the cfg object to remove the member.

Example
To remove mongod_C.example.net:27017 use the following JavaScript operation:
cfg.members.splice(2,1)

6. Overwrite the replica set configuration document with the new configuration by issuing the following:
rs.reconfig(cfg)

As a result of rs.reconfig() the shell will disconnect while the replica set renegotiates which member is
primary. The shell displays a DBClientCursor::init call() failed error even though the command succeeds, and will automatically reconnected.
7. To confirm the new configuration, issue rs.conf().
For the example above the output would be:
{
"_id" : "rs",
"version" : 8,
"members" : [
{
"_id" : 0,
"host" : "mongod_A.example.net:27017"
},
{
"_id" : 1,
"host" : "mongod_B.example.net:27017"
}
]
}

436

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Replace a Replica Set Member

If you need to change the hostname of a replica set member without changing the configuration of that member or the
set, you can use the operation outlined in this tutorial. For example if you must re-provision systems or rename hosts,
you can use this pattern to minimize the scope of that change.
Operation

To change the hostname for a replica set member modify the host (page 468) field. The value of _id (page 467)
field will not change when you reconfigure the set.
See Replica Set Configuration (page 467) and rs.reconfig() for more information.
Note: Any replica set configuration change can trigger the current primary to step down, which forces an election
(page 397). During the election, the current shell session and clients connected to this replica set disconnect, which
produces an error even when the operation succeeds.

Example

To change the hostname to mongo2.example.net for the replica set member configured at members[0], issue
the following sequence of commands:
cfg = rs.conf()
cfg.members[0].host = "mongo2.example.net"
rs.reconfig(cfg)

8.3.2 Member Configuration Tutorials

The following tutorials provide information in configuring replica set members to support specific operations, such as
to provide dedicated backups, to support reporting, or to act as a cold standby.
Adjust Priority for Replica Set Member (page 437) Change the precedence given to a replica set members in an election for primary.
Prevent Secondary from Becoming Primary (page 438) Make a secondary member ineligible for election as primary.
Configure a Hidden Replica Set Member (page 439) Configure a secondary member to be invisible to applications
in order to support significantly different usage, such as a dedicated backups.
Configure a Delayed Replica Set Member (page 441) Configure a secondary member to keep a delayed copy of the
data set in order to provide a rolling backup.
Configure Non-Voting Replica Set Member (page 442) Create a secondary member that keeps a copy of the data set
but does not vote in an election.
Convert a Secondary to an Arbiter (page 443) Convert a secondary to an arbiter.
Adjust Priority for Replica Set Member
To change the value of the priority (page 469) in the replica set configuration, use the following sequence of
commands in the mongo shell:

8.3. Replica Set Tutorials

437

MongoDB Documentation, Release 2.4.9

cfg = rs.conf()
cfg.members[0].priority = 0.5
cfg.members[1].priority = 2
cfg.members[2].priority = 2
rs.reconfig(cfg)

The first operation uses rs.conf() to set the local variable cfg to the contents of the current replica set configuration, which is a document. The next three operations change the priority (page 469) value in the cfg document
for the first three members configured in the members (page 467) array. The final operation calls rs.reconfig()
with the argument of cfg to initialize the new configuration.
When updating the replica configuration object, access the replica set members in the members (page 467) array with
the array index. The array index begins with 0. Do not confuse this index value with the value of the _id (page 467)
field in each document in the members (page 467) array.
If a member has priority (page 469) set to 0, it is ineligible to become primary and will not seek election. Hidden
members (page 387), delayed members (page 388), and arbiters (page ??) all have priority (page 469) set to 0.
All members have a priority (page 469) equal to 1 by default.
The value of priority (page 469) can be any floating point (i.e. decimal) number between 0 and 1000. Priorities
are only used to determine the preference in election. The priority value is used only in relation to other members.
With the exception of members with a priority of 0, the absolute value of the priority (page 469) value is irrelevant.
Replica sets will preferentially elect and maintain the primary status of the member with the highest priority
(page 469) setting.
Warning: Replica set reconfiguration can force the current primary to step down, leading to an election for
primary in the replica set. Elections cause the current primary to close all open client connections.
Perform routine replica set reconfiguration during scheduled maintenance windows.
See also:
The Replica Reconfiguration Usage (page 470) example revolves around changing the priorities of the members
(page 467) of a replica set.
Prevent Secondary from Becoming Primary
To prevent a secondary member from ever becoming a primary in a failover, assign the secondary a priority of 0,
as described here. You can set this secondary-only mode for any member of the replica set, except the current
primary. For a detailed description of secondary-only members and their purposes, see Priority 0 Replica Set Members
(page 386).
To configure a member as secondary-only, set its priority (page 469) value to 0 in the members (page 467)
document in its replica set configuration. Any member with a priority (page 469) equal to 0 will never seek
election (page 397) and cannot become primary in any situation.
{
"_id" : <num>,
"host" : <hostname:port>,
"priority" : 0
}

MongoDB does not permit the current primary to have a priority of 0. To prevent the current primary from again
becoming a primary, you must first step down the current primary using rs.stepDown(), and then you must
reconfigure the replica set (page 470) with rs.conf() and rs.reconfig().

438

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Example

As an example of modifying member priorities, assume a four-member replica set. Use the following sequence of
operations to modify member priorities in the mongo shell connected to the primary. Identify each member by its
array index in the members (page 467) array:
cfg = rs.conf()
cfg.members[0].priority
cfg.members[1].priority
cfg.members[2].priority
cfg.members[3].priority
rs.reconfig(cfg)

=
=
=
=

2
1
0.5
0

The sequence of operations reconfigures the set with the following priority settings:
Member at 0 has a priority of 2 so that it becomes primary under most circumstances.
Member at 1 has a priority of 1, which is the default value. Member 1 becomes primary if no member with a
higher priority is eligible.
Member at 2 has a priority of 0.5, which makes it less likely to become primary than other members but doesnt
prohibit the possibility.
Member at 3 has a priority of 0. Member at 3 cannot become the primary member under any circumstances.
When updating the replica configuration object, access the replica set members in the members (page 467) array with
the array index. The array index begins with 0. Do not confuse this index value with the value of the _id (page 467)
field in each document in the members (page 467) array.
Warning:
The rs.reconfig() shell method can force the current primary to step down, which causes an election
(page 397). When the primary steps down, the mongod closes all client connections. While this typically
takes 10-20 seconds, try to make these changes during scheduled maintenance periods.
To successfully reconfigure a replica set, a majority of the members must be accessible. If your replica set
has an even number of members, add an arbiter (page 431) to ensure that members can quickly obtain a
majority of votes in an election for primary.

Related Documents

priority (page 469)

Adjust Priority for Replica Set Member (page 437)
Replica Set Reconfiguration (page 470)
Replica Set Elections (page 397)
Configure a Hidden Replica Set Member
Hidden members are part of a replica set but cannot become primary and are invisible to client applications. Hidden
members do vote in elections (page 397). For a more information on hidden members and their uses, see Hidden
Replica Set Members (page 387).

8.3. Replica Set Tutorials

439

MongoDB Documentation, Release 2.4.9

Considerations

The most common use of hidden nodes is to support delayed members (page 388). If you only need to prevent a
member from becoming primary, configure a priority 0 member.
If the chainingAllowed (page 470) setting allows secondary members to sync from other secondaries, MongoDB
by default prefers non-hidden members over hidden members when selecting a sync target. MongoDB will only choose
hidden members as a last resort. If you want a secondary to sync from a hidden member, use the replSetSyncFrom
database command to override the default sync target. See the documentation for replSetSyncFrom before using
the command.
See also:
Manage Chained Replication (page 456)
Changed in version 2.0: For sharded clusters running with replica sets before 2.0, if you reconfigured a member as
hidden, you had to restart mongos to prevent queries from reaching the hidden member.
Examples

Member Configuration Document To configure a secondary member as hidden, set its priority (page 469)
value to 0 and set its hidden (page 468) value to true in its member configuration:
{
"_id" : <num>
"host" : <hostname:port>,
"priority" : 0,
"hidden" : true
}

Configuration Procedure The following example hides the secondary member currently at the index 0 in the
members (page 467) array. To configure a hidden member, use the following sequence of operations in a mongo
shell connected to the primary, specifying the member to configure by its array index in the members (page 467)
array:
cfg = rs.conf()
cfg.members[0].priority = 0
cfg.members[0].hidden = true
rs.reconfig(cfg)

After re-configuring the set, this secondary member has a priority of 0 so that it cannot become primary and is hidden.
The other members in the set will not advertise the hidden member in the isMaster or db.isMaster() output.
When updating the replica configuration object, access the replica set members in the members (page 467) array with
the array index. The array index begins with 0. Do not confuse this index value with the value of the _id (page 467)
field in each document in the members (page 467) array.
Warning:
The rs.reconfig() shell method can force the current primary to step down, which causes an election
(page 397). When the primary steps down, the mongod closes all client connections. While this typically
takes 10-20 seconds, try to make these changes during scheduled maintenance periods.
To successfully reconfigure a replica set, a majority of the members must be accessible. If your replica set
has an even number of members, add an arbiter (page 431) to ensure that members can quickly obtain a
majority of votes in an election for primary.

440

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Related Documents

Replica Set Reconfiguration (page 470)

Replica Set Elections (page 397)
Read Preference (page 406)
Configure a Delayed Replica Set Member
To configure a delayed secondary member, set its priority (page 469) value to 0, its hidden (page 468) value to
true, and its slaveDelay (page 469) value to the number of seconds to delay.
Important: The length of the secondary slaveDelay (page 469) must fit within the window of the oplog. If
the oplog is shorter than the slaveDelay (page 469) window, the delayed member cannot successfully replicate
operations.
When you configure a delayed member, the delay applies both to replication and to the members oplog. For details
on delayed members and their uses, see Delayed Replica Set Members (page 388).
Example

The following example sets a 1-hour delay on a secondary member currently at the index 0 in the members (page 467)
array. To set the delay, issue the following sequence of operations in a mongo shell connected to the primary:
cfg = rs.conf()
cfg.members[0].priority = 0
cfg.members[0].hidden = true
cfg.members[0].slaveDelay = 3600
rs.reconfig(cfg)

After the replica set reconfigures, the delayed secondary member cannot become primary and is hidden from applications. The slaveDelay (page 469) value delays both replication and the members oplog by 3600 seconds (1
hour).
When updating the replica configuration object, access the replica set members in the members (page 467) array with
the array index. The array index begins with 0. Do not confuse this index value with the value of the _id (page 467)
field in each document in the members (page 467) array.
Warning:
The rs.reconfig() shell method can force the current primary to step down, which causes an election
(page 397). When the primary steps down, the mongod closes all client connections. While this typically
takes 10-20 seconds, try to make these changes during scheduled maintenance periods.
To successfully reconfigure a replica set, a majority of the members must be accessible. If your replica set
has an even number of members, add an arbiter (page 431) to ensure that members can quickly obtain a
majority of votes in an election for primary.

Related Documents

slaveDelay (page 469)

Replica Set Reconfiguration (page 470)
Oplog Size (page 411)
8.3. Replica Set Tutorials

441

MongoDB Documentation, Release 2.4.9

Change the Size of the Oplog (page 445) tutorial

Replica Set Elections (page 397)
Configure Non-Voting Replica Set Member
Non-voting members allow you to add additional members for read distribution beyond the maximum seven voting
members. To configure a member as non-voting, set its votes (page 469) value to 0.
Example

To disable the ability to vote in elections for the fourth, fifth, and sixth replica set members, use the following command
sequence in the mongo shell connected to the primary. You identify each replica set member by its array index in the
members (page 467) array:
cfg = rs.conf()
cfg.members[3].votes = 0
cfg.members[4].votes = 0
cfg.members[5].votes = 0
rs.reconfig(cfg)

This sequence gives 0 votes to the fourth, fifth, and sixth members of the set according to the order of the members
(page 467) array in the output of rs.conf(). This setting allows the set to elect these members as primary but does
not allow them to vote in elections. Place voting members so that your designated primary or primaries can reach a
majority of votes in the event of a network partition.
When updating the replica configuration object, access the replica set members in the members (page 467) array with
the array index. The array index begins with 0. Do not confuse this index value with the value of the _id (page 467)
field in each document in the members (page 467) array.
Warning:
The rs.reconfig() shell method can force the current primary to step down, which causes an election
(page 397). When the primary steps down, the mongod closes all client connections. While this typically
takes 10-20 seconds, try to make these changes during scheduled maintenance periods.
To successfully reconfigure a replica set, a majority of the members must be accessible. If your replica set
has an even number of members, add an arbiter (page 431) to ensure that members can quickly obtain a
majority of votes in an election for primary.
In general and when possible, all members should have only 1 vote. This prevents intermittent ties, deadlocks, or the
wrong members from becoming primary. Use priority (page 469) to control which members are more likely to
become primary.
Related Documents

votes (page 469)

Replica Set Reconfiguration (page 470)
Replica Set Elections (page 397)

442

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Convert a Secondary to an Arbiter

If you have a secondary in a replica set that no longer needs to hold data but that needs to remain in the set to ensure that
the set can elect a primary (page 397), you may convert the secondary to an arbiter (page ??) using either procedure
in this tutorial. Both procedures are operationally equivalent:
You may operate the arbiter on the same port as the former secondary. In this procedure, you must shut down
the secondary and remove its data before restarting and reconfiguring it as an arbiter.
For this procedure, see Convert Secondary to Arbiter and Reuse the Port Number (page 443).
Run the arbiter on a new port. In this procedure, you can reconfigure the server as an arbiter before shutting
down the instance running as a secondary.
For this procedure, see Convert Secondary to Arbiter Running on a New Port Number (page 444).
Convert Secondary to Arbiter and Reuse the Port Number

1. If your application is connecting directly to the secondary, modify the application so that MongoDB queries
dont reach the secondary.
2. Shut down the secondary.
3. Remove the secondary from the replica set by calling the rs.remove() method. Perform this operation while
connected to the current primary in the mongo shell:
rs.remove("<hostname><:port>")

4. Verify that the replica set no longer includes the secondary by calling the rs.conf() method in the mongo
shell:
rs.conf()

5. Move the secondarys data directory to an archive folder. For example:

mv /data/db /data/db-old

Optional
You may remove the data instead.
6. Create a new, empty data directory to point to when restarting the mongod instance. You can reuse the previous
name. For example:
mkdir /data/db

7. Restart the mongod instance for the secondary, specifying the port number, the empty data directory, and the
replica set. You can use the same port number you used before. Issue a command similar to the following:
mongod --port 27021 --dbpath /data/db --replSet rs

8. In the mongo shell convert the secondary to an arbiter using the rs.addArb() method:
rs.addArb("<hostname><:port>")

9. Verify the arbiter belongs to the replica set by calling the rs.conf() method in the mongo shell.
rs.conf()

The arbiter member should include the following:

8.3. Replica Set Tutorials

443

MongoDB Documentation, Release 2.4.9

"arbiterOnly" : true

Convert Secondary to Arbiter Running on a New Port Number

1. If your application is connecting directly to the secondary or has a connection string referencing the secondary,
modify the application so that MongoDB queries dont reach the secondary.
2. Create a new, empty data directory to be used with the new port number. For example:
mkdir /data/db-temp

3. Start a new mongod instance on the new port number, specifying the new data directory and the existing replica
set. Issue a command similar to the following:
mongod --port 27021 --dbpath /data/db-temp --replSet rs

4. In the mongo shell connected to the current primary, convert the new mongod instance to an arbiter using the
rs.addArb() method:
rs.addArb("<hostname><:port>")

5. Verify the arbiter has been added to the replica set by calling the rs.conf() method in the mongo shell.
rs.conf()

The arbiter member should include the following:

"arbiterOnly" : true

6. Shut down the secondary.

7. Remove the secondary from the replica set by calling the rs.remove() method in the mongo shell:
rs.remove("<hostname><:port>")

8. Verify that the replica set no longer includes the old secondary by calling the rs.conf() method in the mongo
shell:
rs.conf()

9. Move the secondarys data directory to an archive folder. For example:

mv /data/db /data/db-old

Optional
You may remove the data instead.

8.3.3 Replica Set Maintenance Tutorials

The following tutorials provide information in maintaining existing replica sets.
Change the Size of the Oplog (page 445) Increase the size of the oplog which logs operations. In most cases, the
default oplog size is sufficient.
Force a Member to Become Primary (page 447) Force a replica set member to become primary.

444

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Resync a Member of a Replica Set (page 449) Sync the data on a member. Either perform initial sync on a new
member or resync the data on an existing member that has fallen too far behind to catch up by way of normal
replication.
Configure Replica Set Tag Sets (page 450) Assign tags to replica set members for use in targeting read and write
operations to specific members.
Reconfigure a Replica Set with Unavailable Members (page 454) Reconfigure a replica set when a majority of
replica set members are down or unreachable.
Manage Chained Replication (page 456) Disable or enable chained replication. Chained replication occurs when a
secondary replicates from another secondary instead of the primary.
Change Hostnames in a Replica Set (page 457) Update the replica set configuration to reflect changes in members
hostnames.
Configure a Secondarys Sync Target (page 461) Specify the member that a secondary member synchronizes from.
Change the Size of the Oplog
The oplog exists internally as a capped collection, so you cannot modify its size in the course of normal operations. In
most cases the default oplog size (page 411) is an acceptable size; however, in some situations you may need a larger
or smaller oplog. For example, you might need to change the oplog size if your applications perform large numbers of
multi-updates or deletes in short periods of time.
This tutorial describes how to resize the oplog. For a detailed explanation of oplog sizing, see Oplog Size (page 411).
For details how oplog size affects delayed members and affects replication lag, see Delayed Replica Set Members
(page 388).
Overview

To change the size of the oplog, you must perform maintenance on each member of the replica set in turn. The
procedure requires: stopping the mongod instance and starting as a standalone instance, modifying the oplog size,
and restarting the member.
Important: Always start rolling replica set maintenance with the secondaries, and finish with the maintenance on
primary member.

Procedure

Restart the member in standalone mode.

Tip
Always use rs.stepDown() to force the primary to become a secondary, before stopping the server. This
facilitates a more efficient election process.
Recreate the oplog with the new size and with an old oplog entry as a seed.
Restart the mongod instance as a member of the replica set.

8.3. Replica Set Tutorials

445

MongoDB Documentation, Release 2.4.9

Restart a Secondary in Standalone Mode on a Different Port Shut down the mongod instance for one of the
non-primary members of your replica set. For example, to shut down, use the db.shutdownServer() method:
db.shutdownServer()

Restart this mongod as a standalone instance running on a different port and without the --replSet parameter. Use
a command similar to the following:
mongod --port 37017 --dbpath /srv/mongodb

Create a Backup of the Oplog (Optional) Optionally, backup the existing oplog on the standalone instance, as in
the following example:
mongodump --db local --collection 'oplog.rs' --port 37017

Recreate the Oplog with a New Size and a Seed Entry Save the last entry from the oplog. For example, connect
to the instance using the mongo shell, and enter the following command to switch to the local database:
use local

In mongo shell scripts you can use the following operation to set the db object:
db = db.getSiblingDB('local')

Use the db.collection.save() method and a sort on reverse natural order to find the last entry and save it to a
temporary collection:
db.temp.save( db.oplog.rs.find( { }, { ts: 1, h: 1 } ).sort( {$natural : -1} ).limit(1).next() )

To see this oplog entry, use the following operation:

db.temp.find()

Remove the Existing Oplog Collection Drop the old oplog.rs collection in the local database. Use the following command:
db = db.getSiblingDB('local')
db.oplog.rs.drop()

This returns true in the shell.

Create a New Oplog Use the create command to create a new oplog of a different size. Specify the size
argument in bytes. A value of 2 * 1024 * 1024 * 1024 will create a new oplog thats 2 gigabytes:
db.runCommand( { create: "oplog.rs", capped: true, size: (2 * 1024 * 1024 * 1024) } )

Upon success, this command returns the following status:

{ "ok" : 1 }

Insert the Last Entry of the Old Oplog into the New Oplog Insert the previously saved last entry from the old
oplog into the new oplog. For example:
db.oplog.rs.save( db.temp.findOne() )

446

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

To confirm the entry is in the new oplog, use the following operation:
db.oplog.rs.find()

Restart the Member Restart the mongod as a member of the replica set on its usual port. For example:
db.shutdownServer()
mongod --replSet rs0 --dbpath /srv/mongodb

The replica set member will recover and catch up before it is eligible for election to primary.
Repeat Process for all Members that may become Primary Repeat this procedure for all members you want to
change the size of the oplog. Repeat the procedure for the primary as part of the following step.
Change the Size of the Oplog on the Primary To finish the rolling maintenance operation, step down the primary
with the rs.stepDown() method and repeat the oplog resizing procedure above.
Force a Member to Become Primary
Synopsis

You can force a replica set member to become primary by giving it a higher priority (page 469) value than any
other member in the set.
Optionally, you also can force a member never to become primary by setting its priority (page 469) value to 0,
which means the member can never seek election (page 397) as primary. For more information, see Priority 0 Replica
Set Members (page 386).
Procedures

Force a Member to be Primary by Setting its Priority High Changed in version 2.0.
For more information on priorities, see priority (page 469).
This procedure assumes your current primary is m1.example.net and that youd like to instead make
m3.example.net primary. The procedure also assumes you have a three-member replica set with the configuration below. For more information on configurations, see Replica Set Configuration Use (page 470).
This procedure assumes this configuration:
{
"_id" : "rs",
"version" : 7,
"members" : [
{
"_id" : 0,
"host" : "m1.example.net:27017"
},
{
"_id" : 1,
"host" : "m2.example.net:27017"
},
{
"_id" : 2,

8.3. Replica Set Tutorials

447

MongoDB Documentation, Release 2.4.9

"host" : "m3.example.net:27017"
}
]
}

1. In the mongo shell, use the following sequence of operations to make m3.example.net the primary:
cfg = rs.conf()
cfg.members[0].priority = 0.5
cfg.members[1].priority = 0.5
cfg.members[2].priority = 1
rs.reconfig(cfg)

This sets m3.example.net to have a higher local.system.replset.members[n].priority

(page 469) value than the other mongod instances.
The following sequence of events occur:
m3.example.net and m2.example.net sync with m1.example.net (typically within 10 seconds).
m1.example.net sees that it no longer has highest priority and, in most cases, steps down.
m1.example.net does not step down if m3.example.nets sync is far behind. In that case,
m1.example.net waits until m3.example.net is within 10 seconds of its optime and then steps
down. This minimizes the amount of time with no primary following failover.
The step down forces on election in which m3.example.net becomes primary based on its priority
(page 469) setting.
2. Optionally, if m3.example.net is more than 10 seconds behind m1.example.nets optime, and if you
dont need to have a primary designated within 10 seconds, you can force m1.example.net to step down by
running:
db.adminCommand({replSetStepDown: 86400, force: 1})

This prevents m1.example.net from being primary for 86,400 seconds (24 hours), even if there is no other
member that can become primary. When m3.example.net catches up with m1.example.net it will
become primary.
If you later want to make m1.example.net primary again while it waits for m3.example.net to catch
up, issue the following command to make m1.example.net seek election again:
rs.freeze()

The rs.freeze() provides a wrapper around the replSetFreeze database command.

Force a Member to be Primary Using Database Commands Changed in version 1.8.
Consider a replica set with the following members:
mdb0.example.net - the current primary.
mdb1.example.net - a secondary.
mdb2.example.net - a secondary .
To force a member to become primary use the following procedure:
1. In a mongo shell, run rs.status() to ensure your replica set is running as expected.
2. In a mongo shell connected to the mongod instance running on mdb2.example.net, freeze
mdb2.example.net so that it does not attempt to become primary for 120 seconds.

448

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

rs.freeze(120)

3. In a mongo shell connected the mongod running on mdb0.example.net, step down this instance that the
mongod is not eligible to become primary for 120 seconds:
rs.stepDown(120)

mdb1.example.net becomes primary.

Note: During the transition, there is a short window where the set does not have a primary.
For more information, consider the rs.freeze() and rs.stepDown() methods that wrap the
replSetFreeze and replSetStepDown commands.
Resync a Member of a Replica Set
A replica set member becomes stale when its replication process falls so far behind that the primary overwrites
oplog entries the member has not yet replicated. The member cannot catch up and becomes stale. When this occurs,
you must completely resynchronize the member by removing its data and performing an initial sync (page 412).
This tutorial addressed both resyncing a stale member and to creating a new member using seed data from another
member. When syncing a member, choose a time when the system has the bandwidth to move a large amount of data.
Schedule the synchronization during a time of low usage or during a maintenance window.
MongoDB provides two options for performing an initial sync:
Restart the mongod with an empty data directory and let MongoDBs normal initial syncing feature restore the
data. This is the more simple option but may take longer to replace the data.
See Procedures (page 449).
Restart the machine with a copy of a recent data directory from another member in the replica set. This procedure
can replace the data more quickly but requires more manual steps.
See Sync by Copying Data Files from Another Member (page 450).
Procedures

Automatically Sync a Member

Warning: During initial sync, mongod will remove the content of the dbpath.

This procedure relies on MongoDBs regular process for initial sync (page 412). This will store the current data on the
member. For an overview of MongoDB initial sync process, see the Replication Processes (page 410) section.
If the instance has no data, you can simply follow the Add Members to a Replica Set (page 433) or Replace a Replica
Set Member (page 437) procedure to add a new member to a replica set.
You can also force a mongod that is already a member of the set to to perform an initial sync by restarting the instance
without the content of the dbpath as follows:
1. Stop the members mongod instance. To ensure a clean shutdown, use the db.shutdownServer() method
from the mongo shell or on Linux systems, the mongod --shutdown option.
2. Delete all data and sub-directories from the members data directory. By removing the data dbpath, MongoDB
will perform a complete resync. Consider making a backup first.

8.3. Replica Set Tutorials

449

MongoDB Documentation, Release 2.4.9

At this point, the mongod will perform an initial sync. The length of the initial sync process depends on the size of
the database and network connection between members of the replica set.
Initial sync operations can impact the other members of the set and create additional traffic to the primary and can only
occur if another member of the set is accessible and up to date.
Sync by Copying Data Files from Another Member This approach seeds a new or stale member using the data
files from an existing member of the replica set. The data files must be sufficiently recent to allow the new member to
catch up with the oplog. Otherwise the member would need to perform an initial sync.
Copy the Data Files You can capture the data files as either a snapshot or a direct copy. However, in most cases you
cannot copy data files from a running mongod instance to another because the data files will change during the file
copy operation.
Important: If copying data files, you must copy the content of the local database.
You cannot use a mongodump backup to for the data files, only a snapshot backup. For approaches to capture a
consistent snapshot of a running mongod instance, see the MongoDB Backup Methods (page 136) documentation.
Sync the Member After you have copied the data files from the seed source, start the mongod instance and allow
it to apply all operations from the oplog until it reflects the current state of the replica set.
Configure Replica Set Tag Sets
Tag sets let you customize write concern and read preferences for a replica set. MongoDB stores tag sets in the replica
set configuration object, which is the document returned by rs.conf(), in the members[n].tags (page 469)
sub-document.
This section introduces the configuration of tag sets. For an overview on tag sets and their use, see Replica Set Write
Concern (page 49) and Tag Sets (page 408).
Differences Between Read Preferences and Write Concerns

Custom read preferences and write concerns evaluate tags sets in different ways:
Read preferences consider the value of a tag when selecting a member to read from.
Write concerns do not use the value of a tag to select a member except to consider whether or not the value is
unique.
For example, a tag set for a read operation may resemble the following document:
{ "disk": "ssd", "use": "reporting" }

To fulfill such a read operation, a member would need to have both of these tags. Any of the following tag sets would
satisfy this requirement:
{
{
{
{

"disk":
"disk":
"disk":
"disk":

"ssd",
"ssd",
"ssd",
"ssd",

"use":
"use":
"use":
"use":

"reporting" }
"reporting", "rack": "a" }
"reporting", "rack": "d" }
"reporting", "mem": "r"}

The following tag sets would not be able to fulfill this query:

450

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

{
{
{
{
{

"disk": "ssd" }
"use": "reporting" }
"disk": "ssd", "use": "production" }
"disk": "ssd", "use": "production", "rack": "k" }
"disk": "spinning", "use": "reporting", "mem": "32" }

Add Tag Sets to a Replica Set

Given the following replica set configuration:

{
"_id" : "rs0",
"version" : 1,
"members" : [
{
"_id" : 0,
"host" : "mongodb0.example.net:27017"
},
{
"_id" : 1,
"host" : "mongodb1.example.net:27017"
},
{
"_id" : 2,
"host" : "mongodb2.example.net:27017"
}
]
}

You could add tag sets to the members of this replica set with the following command sequence in the mongo shell:
conf = rs.conf()
conf.members[0].tags = { "dc": "east", "use": "production" }
conf.members[1].tags = { "dc": "east", "use": "reporting" }
conf.members[2].tags = { "use": "production" }
rs.reconfig(conf)

After this operation the output of rs.conf() would resemble the following:
{
"_id" : "rs0",
"version" : 2,
"members" : [
{
"_id" : 0,
"host" : "mongodb0.example.net:27017",
"tags" : {
"dc": "east",
"use": "production"
}
},
{
"_id" : 1,
"host" : "mongodb1.example.net:27017",
"tags" : {
"dc": "east",
"use": "reporting"

8.3. Replica Set Tutorials

451

MongoDB Documentation, Release 2.4.9

}
},
{
"_id" : 2,
"host" : "mongodb2.example.net:27017",
"tags" : {
"use": "production"
}
}
]
}

Important: In tag sets, all tag values must be strings.

Custom Multi-Datacenter Write Concerns

Given a five member replica set with members in two data centers:
1. a facility VA tagged dc.va
2. a facility GTO tagged dc.gto
Create a custom write concern to require confirmation from two data centers using replica set tags, using the following
sequence of operations in the mongo shell:
1. Create a replica set configuration JavaScript object conf:
conf = rs.conf()

2. Add tags to the replica set members reflecting their locations:

conf.members[0].tags
conf.members[1].tags
conf.members[2].tags
conf.members[3].tags
conf.members[4].tags
rs.reconfig(conf)

=
=
=
=
=

{
{
{
{
{

"dc.va": "rack1"}
"dc.va": "rack2"}
"dc.gto": "rack1"}
"dc.gto": "rack2"}
"dc.va": "rack1"}

3. Create a custom getLastErrorModes (page 470) setting to ensure that the write operation will propagate
to at least one member of each facility:
conf.settings = { getLastErrorModes: { MultipleDC : { "dc.va": 1, "dc.gto": 1}}

4. Reconfigure the replica set using the modified conf configuration object:
rs.reconfig(conf)

To ensure that a write operation propagates to at least one member of the set in both data centers, use the MultipleDC
write concern mode as follows:
db.runCommand( { getLastError: 1, w: "MultipleDC" } )

Alternatively, if you want to ensure that each write operation propagates to at least 2 racks in each facility, reconfigure
the replica set as follows in the mongo shell:
1. Create a replica set configuration object conf:
conf = rs.conf()

452

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

2. Redefine the getLastErrorModes (page 470) value to require two different values of both dc.va and
dc.gto:
conf.settings = { getLastErrorModes: { MultipleDC : { "dc.va": 2, "dc.gto": 2}}

3. Reconfigure the replica set using the modified conf configuration object:
rs.reconfig(conf)

Now, the following write concern operation will only return after the write operation propagates to at least two different
racks in the each facility:
db.runCommand( { getLastError: 1, w: "MultipleDC" } )

Configure Tag Sets for Functional Segregation of Read and Write Operations

Given a replica set with tag sets that reflect:

data center facility,
physical rack location of instance, and
storage system (i.e. disk) type.
Where each member of the set has a tag set that resembles one of the following:

11

{"dc.va": "rack1", disk:"ssd", ssd: "installed" }

{"dc.va": "rack2", disk:"raid"}
{"dc.gto": "rack1", disk:"ssd", ssd: "installed" }
{"dc.gto": "rack2", disk:"raid"}
{"dc.va": "rack1", disk:"ssd", ssd: "installed" }

To target a read operation to a member of the replica set with a disk type of ssd, you could use the following tag set:
{ disk: "ssd" }

However, to create comparable write concern modes, you would specify a different set of getLastErrorModes
(page 470) configuration. Consider the following sequence of operations in the mongo shell:
1. Create a replica set configuration object conf:
conf = rs.conf()

2. Redefine the getLastErrorModes (page 470) value to configure two write concern modes:
conf.settings = {
"getLastErrorModes" : {
"ssd" : {
"ssd" : 1
},
"MultipleDC" : {
"dc.va" : 1,
"dc.gto" : 1
}
}
}

3. Reconfigure the replica set using the modified conf configuration object:
11

Since read preferences and write concerns use the value of fields in tag sets differently, larger deployments may have some redundancy.

8.3. Replica Set Tutorials

453

MongoDB Documentation, Release 2.4.9

rs.reconfig(conf)

Now you can specify the MultipleDC write concern mode, as in the following operation, to ensure that a write
operation propagates to each data center.
db.runCommand( { getLastError: 1, w: "MultipleDC" } )

Additionally, you can specify the ssd write concern mode to ensure that a write operation propagates to at least one
instance with an SSD.
Reconfigure a Replica Set with Unavailable Members
To reconfigure a replica set when a minority of members are unavailable, use the rs.reconfig() operation on
the current primary, following the example in the Replica Set Reconfiguration Procedure (page 470).
This document provides the following options for re-configuring a replica set when a majority of members are not
accessible:
Reconfigure by Forcing the Reconfiguration (page 454)
Reconfigure by Replacing the Replica Set (page 455)
You may need to use one of these procedures, for example, in a geographically distributed replica set, where no local
group of members can reach a majority. See Replica Set Elections (page 397) for more information on this situation.
Reconfigure by Forcing the Reconfiguration

Changed in version 2.0.

This procedure lets you recover while a majority of replica set members are down or unreachable. You connect to any
surviving member and use the force option to the rs.reconfig() method.
The force option forces a new configuration onto the. Use this procedure only to recover from catastrophic interruptions. Do not use force every time you reconfigure. Also, do not use the force option in any automatic scripts
and do not use force when there is still a primary.
To force reconfiguration:
1. Back up a surviving member.
2. Connect to a surviving member and save the current configuration. Consider the following example commands
for saving the configuration:
cfg = rs.conf()
printjson(cfg)

3. On the same member, remove the down and unreachable members of the replica set from the members
(page 467) array by setting the array equal to the surviving members alone. Consider the following example,
which uses the cfg variable created in the previous step:
cfg.members = [cfg.members[0] , cfg.members[4] , cfg.members[7]]

4. On the same member, reconfigure the set by using the rs.reconfig() command with the force option set
to true:
rs.reconfig(cfg, {force : true})

454

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

This operation forces the secondary to use the new configuration. The configuration is then propagated to all the
surviving members listed in the members array. The replica set then elects a new primary.
Note: When you use force : true, the version number in the replica set configuration increases significantly, by tens or hundreds of thousands. This is normal and designed to prevent set version collisions if you
accidentally force re-configurations on both sides of a network partition and then the network partitioning ends.
5. If the failure or partition was only temporary, shut down or decommission the removed members as soon as
possible.
Reconfigure by Replacing the Replica Set

Use the following procedure only for versions of MongoDB prior to version 2.0. If youre running MongoDB 2.0 or
later, use the above procedure, Reconfigure by Forcing the Reconfiguration (page 454).
These procedures are for situations where a majority of the replica set members are down or unreachable. If a majority
is running, then skip these procedures and instead use the rs.reconfig() command according to the examples in
Example Reconfiguration Operations (page 470).
If you run a pre-2.0 version and a majority of your replica set is down, you have the two options described here. Both
involve replacing the replica set.
Reconfigure by Turning Off Replication This option replaces the replica set with a standalone server.
1. Stop the surviving mongod instances. To ensure a clean shutdown, use an existing control script or use the
db.shutdownServer() method.
For example, to use the db.shutdownServer() method, connect to the server using the mongo shell and
issue the following sequence of commands:
use admin
db.shutdownServer()

2. Create a backup of the data directory (i.e. dbpath) of the surviving members of the set.
Optional
If you have a backup of the database you may instead remove this data.
3. Restart one of the mongod instances without the --replSet parameter.
The data is now accessible and provided by a single server that is not a replica set member. Clients can use this
server for both reads and writes.
When possible, re-deploy a replica set to provide redundancy and to protect your deployment from operational interruption.
Reconfigure by Breaking the Mirror This option selects a surviving replica set member to be the new primary
and to seed a new replica set. In the following procedure, the new primary is db0.example.net. MongoDB
copies the data from db0.example.net to all the other members.
1. Stop the surviving mongod instances. To ensure a clean shutdown, use an existing control script or use the
db.shutdownServer() method.
For example, to use the db.shutdownServer() method, connect to the server using the mongo shell and
issue the following sequence of commands:

8.3. Replica Set Tutorials

455

MongoDB Documentation, Release 2.4.9

use admin
db.shutdownServer()

2. Move the data directories (i.e. dbpath) for all the members except db0.example.net, so that all the
members except db0.example.net have empty data directories. For example:
mv /data/db /data/db-old

3. Move the data files for local database (i.e. local.*) so that db0.example.net has no local database.
For example
mkdir /data/local-old
mv /data/db/local* /data/local-old/

4. Start each member of the replica set normally.

5. Connect to db0.example.net in a mongo shell and run rs.initiate() to initiate the replica set.
6. Add the other set members using rs.add(). For example, to add a member running on db1.example.net
at port 27017, issue the following command:
rs.add("db1.example.net:27017")

MongoDB performs an initial sync on the added members by copying all data from db0.example.net to
the added members.
See also:
Resync a Member of a Replica Set (page 449)
Manage Chained Replication
Starting in version 2.0, MongoDB supports chained replication. A chained replication occurs when a secondary
member replicates from another secondary member instead of from the primary. This might be the case, for example,
if a secondary selects its replication target based on ping time and if the closest member is another secondary.
Chained replication can reduce load on the primary. But chained replication can also result in increased replication
lag, depending on the topology of the network.
New in version 2.2.2.
You can use the chainingAllowed (page 470) setting in Replica Set Configuration (page 467) to disable chained
replication for situations where chained replication is causing lag.
MongoDB enables chained replication by default. This procedure describes how to disable it and how to re-enable it.
Note: If chained replication is disabled, you still can use replSetSyncFrom to specify that a secondary replicates
from another secondary. But that configuration will last only until the secondary recalculates which member to sync
from.

Disable Chained Replication

To disable chained replication, set the chainingAllowed (page 470) field in Replica Set Configuration (page 467)
to false.
You can use the following sequence of commands to set chainingAllowed (page 470) to false:
1. Copy the configuration settings into the cfg object:

456

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

cfg = rs.config()

2. Take note of whether the current configuration settings contain the settings sub-document. If they do, skip
this step.
Warning:
document.

To avoid data loss, skip this step if the configuration settings contain the settings sub-

If the current configuration settings do not contain the settings sub-document, create the sub-document by
issuing the following command:
cfg.settings = { }

3. Issue the following sequence of commands to set chainingAllowed (page 470) to false:
cfg.settings.chainingAllowed = false
rs.reconfig(cfg)

Re-enable Chained Replication

To re-enable chained replication, set chainingAllowed (page 470) to true. You can use the following sequence
of commands:
cfg = rs.config()
cfg.settings.chainingAllowed = true
rs.reconfig(cfg)

Change Hostnames in a Replica Set

For most replica sets, the hostnames in the host (page 468) field never change. However, if organizational needs
change, you might need to migrate some or all host names.
Note: Always use resolvable hostnames for the value of the host (page 468) field in the replica set configuration to
avoid confusion and complexity.

Overview

This document provides two separate procedures for changing the hostnames in the host (page 468) field. Use either
of the following approaches:
Change hostnames without disrupting availability (page 458). This approach ensures your applications will
always be able to read and write data to the replica set, but the approach can take a long time and may incur
downtime at the application layer.
If you use the first procedure, you must configure your applications to connect to the replica set at both the old
and new locations, which often requires a restart and reconfiguration at the application layer and which may
affect the availability of your applications. Re-configuring applications is beyond the scope of this document.
Stop all members running on the old hostnames at once (page 460). This approach has a shorter maintenance
window, but the replica set will be unavailable during the operation.
See also:

8.3. Replica Set Tutorials

457

MongoDB Documentation, Release 2.4.9

Replica Set Reconfiguration Process (page 470), Deploy a Replica Set (page 420), and Add Members to a Replica Set
(page 433).
Assumptions

Given a replica set with three members:

database0.example.com:27017 (the primary)
database1.example.com:27017
database2.example.com:27017
And with the following rs.conf() output:
{
"_id" : "rs",
"version" : 3,
"members" : [
{
"_id" : 0,
"host" : "database0.example.com:27017"
},
{
"_id" : 1,
"host" : "database1.example.com:27017"
},
{
"_id" : 2,
"host" : "database2.example.com:27017"
}
]
}

The following procedures change the members hostnames as follows:

mongodb0.example.net:27017 (the primary)
mongodb1.example.net:27017
mongodb2.example.net:27017
Use the most appropriate procedure for your deployment.
Change Hostnames while Maintaining Replica Set Availability

This procedure uses the above assumptions (page 458).

1. For each secondary in the replica set, perform the following sequence of operations:
(a) Stop the secondary.
(b) Restart the secondary at the new location.
(c) Open a mongo shell connected to the replica sets primary. In our example, the primary runs on port
27017 so you would issue the following command:
mongo --port 27017

458

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

(d) Use rs.reconfig() to update the replica set configuration document (page 467) with the new hostname.
For example, the following sequence of commands updates the hostname for the secondary at the array
index 1 of the members array (i.e. members[1]) in the replica set configuration document:
cfg = rs.conf()
cfg.members[1].host = "mongodb1.example.net:27017"
rs.reconfig(cfg)

For more information on updating the configuration document, see Example Reconfiguration Operations
(page 470).
(e) Make sure your client applications are able to access the set at the new location and that the secondary has
a chance to catch up with the other members of the set.
Repeat the above steps for each non-primary member of the set.
2. Open a mongo shell connected to the primary and step down the primary using the rs.stepDown() method:
rs.stepDown()

The replica set elects another member to the become primary.

3. When the step down succeeds, shut down the old primary.
4. Start the mongod instance that will become the new primary in the new location.
5. Connect to the current primary, which was just elected, and update the replica set configuration document
(page 467) with the hostname of the node that is to become the new primary.
For example, if the old primary was at position 0 and the new primarys hostname is
mongodb0.example.net:27017, you would run:
cfg = rs.conf()
cfg.members[0].host = "mongodb0.example.net:27017"
rs.reconfig(cfg)

6. Open a mongo shell connected to the new primary.

7. To confirm the new configuration, call rs.conf() in the mongo shell.
Your output should resemble:
{
"_id" : "rs",
"version" : 4,
"members" : [
{
"_id" : 0,
"host" : "mongodb0.example.net:27017"
},
{
"_id" : 1,
"host" : "mongodb1.example.net:27017"
},
{
"_id" : 2,
"host" : "mongodb2.example.net:27017"
}
]
}

8.3. Replica Set Tutorials

459

MongoDB Documentation, Release 2.4.9

Change All Hostnames at the Same Time

This procedure uses the above assumptions (page 458).

1. Stop all members in the replica set.
2. Restart each member on a different port and without using the --replSet run-time option. Changing the port
number during maintenance prevents clients from connecting to this host while you perform maintenance. Use
the members usual --dbpath, which in this example is /data/db1. Use a command that resembles the
following:
mongod --dbpath /data/db1/ --port 37017

3. For each member of the replica set, perform the following sequence of operations:
(a) Open a mongo shell connected to the mongod running on the new, temporary port. For example, for a
member running on a temporary port of 37017, you would issue this command:
mongo --port 37017

(b) Edit the replica set configuration manually. The replica set configuration is the only document in the
system.replset collection in the local database. Edit the replica set configuration with the new
hostnames and correct ports for all the members of the replica set. Consider the following sequence of
commands to change the hostnames in a three-member set:
use local
cfg = db.system.replset.findOne( { "_id": "rs" } )
cfg.members[0].host = "mongodb0.example.net:27017"
cfg.members[1].host = "mongodb1.example.net:27017"
cfg.members[2].host = "mongodb2.example.net:27017"
db.system.replset.update( { "_id": "rs" } , cfg )

(c) Stop the mongod process on the member.

4. After re-configuring all members of the set, start each mongod instance in the normal way: use the usual port
number and use the --replSet option. For example:
mongod --dbpath /data/db1/ --port 27017 --replSet rs

5. Connect to one of the mongod instances using the mongo shell. For example:
mongo --port 27017

6. To confirm the new configuration, call rs.conf() in the mongo shell.

Your output should resemble:
{
"_id" : "rs",
"version" : 4,
"members" : [
{
"_id" : 0,
"host" : "mongodb0.example.net:27017"
},
{

460

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

"_id" : 1,
"host" : "mongodb1.example.net:27017"
},
{
"_id" : 2,
"host" : "mongodb2.example.net:27017"
}
]
}

Configure a Secondarys Sync Target

To override the default sync target selection logic, you may manually configure a secondary members sync target for
pulling oplog entries temporarily. The following operations provide access to this functionality:
replSetSyncFrom command, or
rs.syncFrom() helper in the mongo shell
Only modify the default sync logic as needed, and always exercise caution. rs.syncFrom() will not affect an inprogress initial sync operation. To affect the sync target for the initial sync, run rs.syncFrom() operation before
initial sync.
If you run rs.syncFrom() during initial sync, MongoDB produces no error messages, but the sync target will not
change until after the initial sync operation.
Note: replSetSyncFrom and rs.syncFrom() provide a temporary override of default behavior. mongod will
revert to the default sync behavior in the following situations:
The mongod instance restarts.
The connection between the mongod and the sync target closes.
Changed in version 2.4: The sync target falls more than 30 seconds behind another member of the replica set; the
mongod will revert to the default sync target.

8.3.4 Troubleshoot Replica Sets

This section describes common strategies for troubleshooting replica set deployments.
Check Replica Set Status
To display the current state of the replica set and current state of each member, run the rs.status() method in a
mongo shell connected to the replica sets primary. For descriptions of the information displayed by rs.status(),
see http://docs.mongodb.org/manualreference/command/replSetGetStatus.
Note: The rs.status() method is a wrapper that runs the replSetGetStatus database command.

Check the Replication Lag

Replication lag is a delay between an operation on the primary and the application of that operation from the oplog to
the secondary. Replication lag can be a significant issue and can seriously affect MongoDB replica set deployments.

8.3. Replica Set Tutorials

461

MongoDB Documentation, Release 2.4.9

Excessive replication lag makes lagged members ineligible to quickly become primary and increases the possibility
that distributed read operations will be inconsistent.
To check the current length of replication lag:
In a mongo shell connected to the primary, call the db.printSlaveReplicationInfo() method.
The returned document displays the syncedTo value for each member, which shows you when each member
last read from the oplog, as shown in the following example:
source:
m1.example.net:30001
syncedTo: Tue Oct 02 2012 11:33:40 GMT-0400 (EDT)
= 7475 secs ago (2.08hrs)
source:
m2.example.net:30002
syncedTo: Tue Oct 02 2012 11:33:40 GMT-0400 (EDT)
= 7475 secs ago (2.08hrs)

Note: The rs.status() method is a wrapper around the replSetGetStatus database command.
Monitor the rate of replication by watching the oplog time in the replica graph in the MongoDB Management
Service12 . For more information see the documentation for MMS13 .
Possible causes of replication lag include:
Network Latency
Check the network routes between the members of your set to ensure that there is no packet loss or network
routing issue.
Use tools including ping to test latency between set members and traceroute to expose the routing of
packets network endpoints.
Disk Throughput
If the file system and disk device on the secondary is unable to flush data to disk as quickly as the primary,
then the secondary will have difficulty keeping state. Disk-related issues are incredibly prevalent on multitenant systems, including vitalized instances, and can be transient if the system accesses disk devices over an IP
network (as is the case with Amazons EBS system.)
Use system-level tools to assess disk status, including iostat or vmstat.
Concurrency
In some cases, long-running operations on the primary can block replication on secondaries. For best results,
configure write concern (page 47) to require confirmation of replication to secondaries, as described in replica
set write concern (page 49). This prevents write operations from returning if replication cannot keep up with
the write load.
Use the database profiler to see if there are slow queries or long-running operations that correspond to the
incidences of lag.
Appropriate Write Concern
If you are performing a large data ingestion or bulk load operation that requires a large number of writes to the
primary, particularly with unacknowledged write concern (page 47), the secondaries will not be able to read the
oplog fast enough to keep up with changes.
To prevent this, require write acknowledgment or journaled write concern (page 47) after every 100, 1,000, or
an another interval to provide an opportunity for secondaries to catch up with the primary.
For more information see:
12 http://mms.mongodb.com/
13 http://mms.mongodb.com/help/

462

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Replica Acknowledge Write Concern (page 49)

Replica Set Write Concern (page 52)
Oplog Size (page 411)
Test Connections Between all Members
All members of a replica set must be able to connect to every other member of the set to support replication. Always verify connections in both directions. Networking topologies and firewall configurations prevent normal and
required connectivity, which can block replication.
Consider the following example of a bidirectional test of networking:
Example
Given a replica set with three members running on three separate hosts:
m1.example.net
m2.example.net
m3.example.net
1. Test the connection from m1.example.net to the other hosts with the following operation set
m1.example.net:
mongo --host m2.example.net --port 27017
mongo --host m3.example.net --port 27017

2. Test the connection from m2.example.net to the other two hosts with the following operation set from
m2.example.net, as in:
mongo --host m1.example.net --port 27017
mongo --host m3.example.net --port 27017

You have now tested the connection between m2.example.net and m1.example.net in both directions.
3. Test the connection from m3.example.net to the other two hosts with the following operation set from the
m3.example.net host, as in:
mongo --host m1.example.net --port 27017
mongo --host m2.example.net --port 27017

If any connection, in any direction fails, check your networking and firewall configuration and reconfigure your environment to allow these connections.

Socket Exceptions when Rebooting More than One Secondary

When you reboot members of a replica set, ensure that the set is able to elect a primary during the maintenance. This
means ensuring that a majority of the sets votes (page 469) are available.
When a sets active members can no longer form a majority, the sets primary steps down and becomes a secondary.
The former primary closes all open connections to client applications. Clients attempting to write to the former primary
receive socket exceptions and Connection reset errors until the set can elect a primary.

8.3. Replica Set Tutorials

463

MongoDB Documentation, Release 2.4.9

Example
Given a three-member replica set where every member has one vote, the set can elect a primary only as long as two
members can connect to each other. If two you reboot the two secondaries once, the primary steps down and becomes
a secondary. Until the at least one secondary becomes available, the set has no primary and cannot elect a new primary.
For more information on votes, see Replica Set Elections (page 397). For related information on connection errors,
see Does TCP keepalive time affect sharded clusters and replica sets? (page 587).
Check the Size of the Oplog
A larger oplog can give a replica set a greater tolerance for lag, and make the set more resilient.
To check the size of the oplog for a given replica set member, connect to the member in a mongo shell and run the
db.printReplicationInfo() method.
The output displays the size of the oplog and the date ranges of the operations contained in the oplog. In the following
example, the oplog is about 10MB and is able to fit about 26 hours (94400 seconds) of operations:
configured oplog size:
log length start to end:
oplog first event time:
oplog last event time:
now:

10.10546875MB
94400 (26.22hrs)
Mon Mar 19 2012 13:50:38 GMT-0400 (EDT)
Wed Oct 03 2012 14:59:10 GMT-0400 (EDT)
Wed Oct 03 2012 15:00:21 GMT-0400 (EDT)

The oplog should be long enough to hold all transactions for the longest downtime you expect on a secondary. At a
minimum, an oplog should be able to hold minimum 24 hours of operations; however, many users prefer to have 72
hours or even a weeks work of operations.
For more information on how oplog size affects operations, see:
Oplog Size (page 411),
Delayed Replica Set Members (page 388), and
Check the Replication Lag (page 461).
Note: You normally want the oplog to be the same size on all members. If you resize the oplog, resize it on all
members.
To change oplog size, see the Change the Size of the Oplog (page 445) tutorial.
Oplog Entry Timestamp Error
Consider the following error in mongod output and logs:
replSet error fatal couldn't query the local local.oplog.rs collection.
<timestamp> [rsStart] bad replSet oplog entry?

Terminating mongod after 30

Often, an incorrectly typed value in the ts field in the last oplog entry causes this error. The correct data type is
Timestamp.
Check the type of the ts value using the following two queries against the oplog collection:
db = db.getSiblingDB("local")
db.oplog.rs.find().sort({$natural:-1}).limit(1)
db.oplog.rs.find({ts:{$type:17}}).sort({$natural:-1}).limit(1)

464

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

The first query returns the last document in the oplog, while the second returns the last document in the oplog where
the ts value is a Timestamp. The $type operator allows you to select BSON type 17, is the Timestamp data type.
If the queries dont return the same document, then the last document in the oplog has the wrong data type in the ts
field.
Example
If the first query returns this as the last oplog entry:
{ "ts" : {t: 1347982456000, i: 1},
"h" : NumberLong("8191276672478122996"),
"op" : "n",
"ns" : "",
"o" : { "msg" : "Reconfig set", "version" : 4 } }

And the second query returns this as the last entry where ts has the Timestamp type:
{ "ts" : Timestamp(1347982454000, 1),
"h" : NumberLong("6188469075153256465"),
"op" : "n",
"ns" : "",
"o" : { "msg" : "Reconfig set", "version" : 3 } }

Then the value for the ts field in the last oplog entry is of the wrong data type.
To set the proper type for this value and resolve this issue, use an update operation that resembles the following:
db.oplog.rs.update( { ts: { t:1347982456000, i:1 } },
{ $set: { ts: new Timestamp(1347982456000, 1)}})

Modify the timestamp values as needed based on your oplog entry. This operation may take some period to complete
because the update must scan and pull the entire oplog into memory.
Duplicate Key Error on local.slaves
The duplicate key on local.slaves error, occurs when a secondary or slave changes its hostname and the primary or
master tries to update its local.slaves collection with the new name. The update fails because it contains the
same _id value as the document containing the previous hostname. The error itself will resemble the following.
exception 11000 E11000 duplicate key error index: local.slaves.$_id_

dup key: { : ObjectId('<object

This is a benign error and does not affect replication operations on the secondary or slave.
To prevent the error from appearing, drop the local.slaves collection from the primary or master, with the
following sequence of operations in the mongo shell:
use local
db.slaves.drop()

The next time a secondary or slave polls the primary or master, the primary or master recreates the local.slaves
collection.

8.3. Replica Set Tutorials

465

MongoDB Documentation, Release 2.4.9

8.4 Replication Reference

8.4.1 Replication Methods in the mongo Shell
Name
Description
rs.add()
Adds a member to a replica set.
rs.addArb()Adds an arbiter to a replica set.
rs.conf() Returns the replica set configuration document.
rs.freeze()Prevents the current member from seeking election as primary for a period of time.
rs.help() Returns basic help text for replica set functions.
rs.initiate()
Initializes a new replica set.
rs.reconfig()
Re-configures a replica set by applying a new replica set configuration object.
rs.remove()Remove a member from a replica set.
rs.slaveOk()
Sets the slaveOk property for the current connection. Deprecated. Use readPref() and
Mongo.setReadPref() to set read preference.
rs.status()Returns a document with information about the state of the replica set.
rs.stepDown()
Causes the current primary to become a secondary which forces an election.
rs.syncFrom()
Sets the member that this replica set member will sync from, overriding the default sync target
selection logic.

8.4.2 Replication Database Commands

Name
Description
replSetFreeze
Prevents the current member from seeking election as primary for a period of time.
replSetGetStatus Returns a document that reports on the status of the replica set.
replSetInitiate
Initializes a new replica set.
replSetMaintenanceEnables or disables a maintenance mode, which puts a secondary node in a
RECOVERING state.
replSetReconfig
Applies a new configuration to an existing replica set.
replSetStepDown
Forces the current primary to step down and become a secondary, forcing an election.
replSetSyncFrom
Explicitly override the default logic for selecting a member to replicate from.
resync
Forces a mongod to re-synchronize from the master. For master-slave replication only.
applyOps
Internal command that applies oplog entries to the current data set.
isMaster
Displays information about this members role in the replica set, including whether it is
the master.
getoptime
Internal command to support replication, returns the optime.

8.4.3 Replica Set Reference Documentation

Replica Set Configuration (page 467) Complete documentation of the replica set configuration object returned by
rs.conf().
The local Database (page 472) Complete documentation of the content of the local database that mongod instances use to support replication.
Replica Set Member States (page 474) Reference for the replica set member states.
Read Preference Reference (page 476) Complete documentation of the five read preference modes that the MongoDB drivers support.

466

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

Replica Set Configuration

Synopsis

This reference provides an overview of replica set configuration options and settings.
Use rs.conf() in the mongo shell to retrieve this configuration. Note that default values are not explicitly displayed.
Example Configuration Document

The following document provides a representation of a replica set configuration document. Angle brackets (e.g. < and
>) enclose all optional fields.
{
_id : <setname>,
version: <int>,
members: [
{
_id : <ordinal>,
host : hostname<:port>,
<arbiterOnly : <boolean>,>
<buildIndexes : <boolean>,>
<hidden : <boolean>,>
<priority: <priority>,>
<tags: { <document> },>
<slaveDelay : <number>,>
<votes : <number>>
}
, ...
],
<settings: {
<getLastErrorDefaults : <lasterrdefaults>,>
<chainingAllowed : <boolean>,>
<getLastErrorModes : <modes>>
}>
}

Configuration Variables

local.system.replset._id
Type: string
Value: <setname>
An _id field holding the name of the replica set. This reflects the set name configured with replSet or
mongod --replSet.
local.system.replset.members
Type: array
Contains an array holding an embedded document for each member of the replica set. The members document
contains a number of fields that describe the configuration of each member of the replica set.
The members (page 467) field in the replica set configuration document is a zero-indexed array.

8.4. Replication Reference

467

MongoDB Documentation, Release 2.4.9

local.system.replset.members[n]._id
Type: ordinal
Provides the zero-indexed identifier of every member in the replica set.
Note: When updating the replica configuration object, access the replica set members in the members
(page 467) array with the array index. The array index begins with 0. Do not confuse this index value with the
value of the _id (page 467) field in each document in the members (page 467) array.
local.system.replset.members[n].host
Type: <hostname><:port>
Identifies the host name of the set member with a hostname and port number. This name must be resolvable for
every host in the replica set.
Warning: host (page 468) cannot hold a value that resolves to localhost or the local interface unless
all members of the set are on hosts that resolve to localhost.
local.system.replset.members[n].arbiterOnly
Optional.
Type: boolean
Default: false
Identifies an arbiter. For arbiters, this value is true, and is automatically configured by rs.addArb().
local.system.replset.members[n].buildIndexes
Optional.
Type: boolean
Default: true
Determines whether the mongod builds indexes on this member. Do not set to false for instances that receive
queries from clients.
Omitting index creation, and thus this setting, may be useful, if:
You are only using this instance to perform backups using mongodump,
this instance will receive no queries, and
index creation and maintenance overburdens the host system.
If set to false, secondaries configured with this option do build indexes on the _id field, to facilitate operations required for replication.
Warning: You may only set this value when adding a member to a replica set. You may not reconfigure a
replica set to change the value of the buildIndexes (page 468) field after adding the member to the set.
buildIndexes (page 468) is only valid when priority is 0 to prevent these members from becoming
primary. Make all instances that do not build indexes hidden.
Other secondaries cannot replicate from a members where buildIndexes (page 468) is false.
local.system.replset.members[n].hidden
Optional.
Type: boolean
Default: false

468

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

When this value is true, the replica set hides this instance, and does not include the member in the output of
db.isMaster() or isMaster. This prevents read operations (i.e. queries) from ever reaching this host by
way of secondary read preference.
See also:
Hidden Replica Set Members (page 387)
local.system.replset.members[n].priority
Optional.
Type: Number, between 0 and 100.0 including decimals.
Default: 1
Specify higher values to make a member more eligible to become primary, and lower values to make the member
less eligible to become primary. Priorities are only used in comparison to each other. Members of the set will
veto election requests from members when another eligible member has a higher priority value. Changing the
balance of priority in a replica set will trigger an election.
A priority (page 469) of 0 makes it impossible for a member to become primary.
See also:
priority (page 469) and Replica Set Elections (page 397).
local.system.replset.members[n].tags
Optional.
Type: MongoDB Document
Default: none
Used to represent arbitrary values for describing or tagging members for the purposes of extending write concern
to allow configurable data center awareness.
Use in conjunction with getLastErrorModes (page 470) and getLastErrorDefaults (page 470) and
db.getLastError() (i.e. getLastError.)
For procedures on configuring tag sets, see Configure Replica Set Tag Sets (page 450).
Important: In tag sets, all tag values must be strings.
local.system.replset.members[n].slaveDelay
Optional.
Type: Integer. (seconds.)
Default: 0
Describes the number of seconds behind the primary that this replica set member should lag. Use this
option to create delayed members (page 388), that maintain a copy of the data that reflects the state of the data
set at some amount of time in the past, specified in seconds. Typically such delayed members help protect
against human error, and provide some measure of insurance against the unforeseen consequences of changes
and updates.
local.system.replset.members[n].votes
Optional.
Type: Integer
Default: 1
Controls the number of votes a server will cast in a replica set election (page 397). The number of votes each
member has can be any non-negative integer, but it is highly recommended each member has 1 or 0 votes.
8.4. Replication Reference

469

MongoDB Documentation, Release 2.4.9

If you need more than 7 members in one replica set, use this setting to add additional non-voting members with
a votes (page 469) value of 0.
For most deployments and most members, use the default value, 1, for votes (page 469).
local.system.replset.settings
Optional.
Type: MongoDB Document
The settings document configures options that apply to the whole replica set.
local.system.replset.settings.chainingAllowed
Optional.
Type: boolean
Default: true
New in version 2.2.4.
When chainingAllowed (page 470) is true, the replica set allows secondary members to replicate from
other secondary members. When chainingAllowed (page 470) is false, secondaries can replicate only
from the primary.
When you run rs.config() to view a replica sets configuration, the chainingAllowed (page 470) field
appears only when set to false. If not set, chainingAllowed (page 470) is true.
See also:
Manage Chained Replication (page 456)
local.system.replset.settings.getLastErrorDefaults
Optional.
Type: MongoDB Document
Specify arguments to getLastError that members of this replica set will use when getLastError has no
arguments. If you specify any arguments, getLastError, ignores these defaults.
local.system.replset.settings.getLastErrorModes
Optional.
Type: MongoDB Document
Defines the names and combination of members (page 467) for use by the application layer to guarantee write
concern to database using the getLastError command to provide data-center awareness.
Example Reconfiguration Operations

Most modifications of replica set configuration use the mongo shell. Consider the following reconfiguration operation:
Example
Given the following replica set configuration:
{
"_id" : "rs0",
"version" : 1,
"members" : [
{
"_id" : 0,
"host" : "mongodb0.example.net:27017"

470

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

},
{
"_id" : 1,
"host" : "mongodb1.example.net:27017"
},
{
"_id" : 2,
"host" : "mongodb2.example.net:27017"
}
]
}

The following reconfiguration operation updates the priority (page 469) of the replica set members:
cfg = rs.conf()
cfg.members[0].priority = 0.5
cfg.members[1].priority = 2
cfg.members[2].priority = 2
rs.reconfig(cfg)

First, this operation sets the local variable cfg to the current replica set configuration using the rs.conf() method.
Then it adds priority values to the cfg document for the three sub-documents in the members (page 467) array,
accessing each replica set member with the array index and not the replica set members _id (page 467) field. Finally,
it calls the rs.reconfig() method with the argument of cfg to initialize this new configuration. The replica set
configuration after this operation will resemble the following:
{
"_id" : "rs0",
"version" : 1,
"members" : [
{
"_id" : 0,
"host" : "mongodb0.example.net:27017",
"priority" : 0.5
},
{
"_id" : 1,
"host" : "mongodb1.example.net:27017",
"priority" : 2
},
{
"_id" : 2,
"host" : "mongodb2.example.net:27017",
"priority" : 1
}
]
}

Using the dot notation demonstrated in the above example, you can modify any existing setting or specify any of
optional replica set configuration variables (page 467). Until you run rs.reconfig(cfg) at the shell, no changes
will take effect. You can issue cfg = rs.conf() at any time before using rs.reconfig() to undo your
changes and start from the current configuration. If you issue cfg as an operation at any point, the mongo shell at
any point will output the complete document with modifications for your review.
The rs.reconfig() operation has a force option, to make it possible to reconfigure a replica set if a majority of
the replica set is not visible, and there is no primary member of the set. use the following form:

8.4. Replication Reference

471

MongoDB Documentation, Release 2.4.9

rs.reconfig(cfg, { force: true } )

Warning: Forcing a rs.reconfig() can lead to rollback situations and other difficult to recover from situations. Exercise caution when using this option.

Note: The rs.reconfig() shell method can force the current primary to step down and triggers an election in
some situations. When the primary steps down, all clients will disconnect. This is by design. Since this typically takes
10-20 seconds, attempt to make such changes during scheduled maintenance periods.

The local Database

Overview

Every mongod instance has its own local database, which stores data used in the replication process, and other
instance-specific data. The local database is invisible to replication: collections in the local database are not
replicated.
In replication, the local database store stores internal replication data for each member of a replica set. The local
stores the following collections:
Changed in version 2.4: When running with authentication (i.e. auth), authenticating to the local database is
not equivalent to authenticating to the admin database. In previous versions, authenticating to the local database
provided access to all databases.
Collection on all mongod Instances

local.startup_log
On startup, each mongod instance inserts a document into startup_log (page 472) with diagnostic information about the mongod instance itself and host information. startup_log (page 472) is a capped collection.
This information is primarily useful for diagnostic purposes.
Example
Consider the following prototype of a document from the startup_log (page 472) collection:
{
"_id" : "<string>",
"hostname" : "<string>",
"startTime" : ISODate("<date>"),
"startTimeLocal" : "<string>",
"cmdLine" : {
"dbpath" : "<path>",
"<option>" : <value>
},
"pid" : <number>,
"buildinfo" : {
"version" : "<string>",
"gitVersion" : "<string>",
"sysInfo" : "<string>",
"loaderFlags" : "<string>",
"compilerFlags" : "<string>",
"allocator" : "<string>",
"versionArray" : [ <num>, <num>, <...> ],

472

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

"javascriptEngine" : "<string>",
"bits" : <number>,
"debug" : <boolean>,
"maxBsonObjectSize" : <number>
}
}

Documents in the startup_log (page 472) collection contain the following fields:
local.startup_log._id
Includes the system hostname and a millisecond epoch value.
local.startup_log.hostname
The systems hostname.
local.startup_log.startTime
A UTC ISODate value that reflects when the server started.
local.startup_log.startTimeLocal
A string that reports the startTime (page 473) in the systems local time zone.
local.startup_log.cmdLine
A sub-document that reports the mongod runtime options and their values.
local.startup_log.pid
The process identifier for this process.
local.startup_log.buildinfo
A sub-document that reports information about the build environment and settings used to compile this
mongod. This is the same output as buildInfo. See buildInfo.

Collections on Replica Set Members

local.system.replset
local.system.replset (page 473) holds the replica sets configuration object as its single document. To
view the objects configuration information, issue rs.conf() from the mongo shell. You can also query this
collection directly.
local.oplog.rs
local.oplog.rs (page 473) is the capped collection that holds the oplog. You set its size at creation using
the oplogSize setting. To resize the oplog after replica set initiation, use the Change the Size of the Oplog
(page 445) procedure. For additional information, see the Oplog Size (page 411) section.
local.replset.minvalid
This contains an object used internally by replica sets to track replication status.
local.slaves
This contains information about each member of the set and the latest point in time that this member has synced
to. If this collection becomes out of date, you can refresh it by dropping the collection and allowing MongoDB
to automatically refresh it during normal replication:
db.getSiblingDB("local").slaves.drop()

Collections used in Master/Slave Replication

In master/slave replication, the local database contains the following collections:

8.4. Replication Reference

473

MongoDB Documentation, Release 2.4.9

On the master:
local.oplog.$main
This is the oplog for the master-slave configuration.
local.slaves
This contains information about each slave.
On each slave:
local.sources
This contains information about the slaves master server.
Replica Set Member States
Members of replica sets have states that reflect the startup process, basic operations, and potential error states.
Number
0
1
2
3
4
5
6
7
8
9
10

Name

State Description

STARTUP
(page 475)
PRIMARY
(page 474)
SECONDARY
(page 474)
RECOVERING
(page 475)
FATAL
(page 475)
STARTUP2
(page 475)
UNKNOWN
(page 475)
ARBITER
(page 474)
DOWN
(page 475)
ROLLBACK
(page 475)
SHUNNED
(page 475)

Cannot vote. All members start up in this state. The mongod parses the replica set
configuration document (page 437) while in STARTUP (page 475).
Can vote. The primary (page 382) is the only member to accept write operations.
Can vote. The secondary (page 382) replicates the data store.
Can vote. Members either perform startup self-checks, or transition from completing a
rollback (page 401) or resync (page 449).
Cannot vote. Has encountered an unrecoverable error.
Cannot vote. Forks replication and election threads before becoming a secondary.
Cannot vote. Has never connected to the replica set.
Can vote. Arbiters (page ??) do not replicate data and exist solely to participate in
elections.
Cannot vote. Is not accessible to the set.
Can vote. Performs a rollback (page 401).
Cannot vote. Was once in the replica set but has now been removed.

States

Core States
PRIMARY
Members in PRIMARY (page 474) state accept write operations. A replica set has only one primary at a time.
A SECONDARY (page 474) member becomes primary after an election (page 397). Members in the PRIMARY
(page 474) state are eligible to vote.
SECONDARY
Members in SECONDARY (page 474) state replicate the primarys data set and can be configured to accept read
operations. Secondaries are eligible to vote in elections, and may be elected to the PRIMARY (page 474) state if
the primary becomes unavailable.

474

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

ARBITER
Members in ARBITER (page 474) state do not replicate data or accept write operations. They are eligible to
vote, and exist solely to break a tie during elections. Replica sets should only have a member in the ARBITER
(page 474) state if the set would otherwise have an even number of members, and could suffer from tied elections. Like primaries, there should only be at most one arbiter in any replica set.
See Replica Set Members (page 382) for more information on core states.
Initialization States
STARTUP
Each member of a replica set starts up in STARTUP (page 475) state. mongod then loads that members
replica set configuration, and transitions the members state to STARTUP2 (page 475). Members in STARTUP
(page 475) are not eligible to vote.
STARTUP2
Each member of a replica set enters the STARTUP2 (page 475) state as soon as mongod finishes loading
that members configuration. While in the STARTUP2 (page 475) state, the member creates threads to handle
internal replication operations. Members are in the STARTUP2 (page 475) state for a short period of time before
entering the RECOVERING (page 475) state. Members in the STARTUP2 (page 475) state are not eligible to
vote.
RECOVERING
A member of a replica set enters RECOVERING (page 475) state when it is not ready to accept reads. The
RECOVERING (page 475) state can occur during normal operation, and doesnt necessarily reflect an error
condition. Members in the RECOVERING (page 475) state are eligible to vote in elections, but is not eligible to
enter the PRIMARY (page 474) state.
During startup, members transition through RECOVERING (page 475) after STARTUP2 (page 475) and before
becoming SECONDARY (page 474).
During normal operation, if a secondary falls behind the other members of the replica set, it may need to resync
(page 449) with the rest of the set. While resyncing, the member enters the RECOVERING (page 475) state.
Whenever the replica set replaces a primary in an election, the old primarys data collection may contain documents that did not have time to replicate to the secondary members. In this case the member rolls back those
writes. During rollback (page 401), the member will have RECOVERING (page 475) state.
On secondaries, the compact and replSetMaintenance commands force the secondary to enter
RECOVERING (page 475) state. This prevents clients from reading during those operations.
Error States Members in any error state cant vote.
FATAL
Members that encounter an unrecoverable error enter the FATAL (page 475) state. Members in this state requires
administrator intervention.
UNKNOWN
Members that have never communicated status information to the replica set are in the UNKNOWN (page 475)
state.
DOWN
Members that lose their connection to the replica set enter the DOWN (page 475) state.
SHUNNED
Members that are removed from the replica set enter the SHUNNED (page 475) state.
ROLLBACK
When a SECONDARY (page 474) rolls back a write operation after transitioning from PRIMARY (page 474), it
enters the ROLLBACK (page 475) state. See Rollbacks During Replica Set Failover (page 401).

8.4. Replication Reference

475

MongoDB Documentation, Release 2.4.9

Read Preference Reference

Read preference describes how MongoDB clients route read operations to members of a replica set.
By default, an application directs its read operations to the primary member in a replica set. Reading from the primary
guarantees that read operations reflect the latest version of a document. However, by distributing some or all reads to
secondary members of the replica set, you can improve read throughput or reduce latency for an application that does
not require fully up-to-date data.
Read Preference
Description
Mode
primary (page 476)
Default mode. All operations read from the current replica set primary.
primaryPreferred In most situations, operations read from the primary but if it is unavailable, operations
(page 476)
read from secondary members.
secondary
All operations read from the secondary members of the replica set.
(page 476)
secondaryPreferred In most situations, operations read from secondary members but if no secondary
(page 477)
members are available, operations read from the primary.
nearest (page 477)
Operations read from member of the replica set with the least network latency,
irrespective of the members type.
Read Preference Modes

primary
All read operations use only the current replica set primary. This is the default. If the primary is unavailable,
read operations produce an error or throw an exception.
The primary (page 476) read preference mode is not compatible with read preference modes that use tag sets
(page 408). If you specify a tag set with primary (page 476), the driver will produce an error.
primaryPreferred
In most situations, operations read from the primary member of the set. However, if the primary is unavailable,
as is the case during failover situations, operations read from secondary members.
When the read preference includes a tag set (page 408), the client reads first from the primary, if available, and
then from secondaries that match the specified tags. If no secondaries have matching tags, the read operation
produces an error.
Since the application may receive data from a secondary, read operations using the primaryPreferred
(page 476) mode may return stale data in some situations.
Warning: Changed in version 2.2: mongos added full support for read preferences.
When connecting to a mongos instance older than 2.2, using a client that supports read preference modes,
primaryPreferred (page 476) will send queries to secondaries.
secondary
Operations read only from the secondary members of the set. If no secondaries are available, then this read
operation produces an error or exception.
Most sets have at least one secondary, but there are situations where there may be no available secondary. For
example, a set with a primary, a secondary, and an arbiter may not have any secondaries if a member is in
recovering state or unavailable.
When the read preference includes a tag set (page 408), the client attempts to find secondary members that
match the specified tag set and directs reads to a random secondary from among the nearest group (page 408).

476

Chapter 8. Replication

MongoDB Documentation, Release 2.4.9

If no secondaries have matching tags, the read operation produces an error.

14

Read operations using the secondary (page 476) mode may return stale data.
secondaryPreferred
In most situations, operations read from secondary members, but in situations where the set consists of a single
primary (and no other members), the read operation will use the sets primary.
When the read preference includes a tag set (page 408), the client attempts to find a secondary member that
matches the specified tag set and directs reads to a random secondary from among the nearest group (page 408).
If no secondaries have matching tags, the client ignores tags and reads from the primary.
Read operations using the secondaryPreferred (page 477) mode may return stale data.
nearest
The driver reads from the nearest member of the set according to the member selection (page 408) process.
Reads in the nearest (page 477) mode do not consider the members type. Reads in nearest (page 477)
mode may read from both primaries and secondaries.
Set this mode to minimize the effect of network latency on read operations without preference for current or
stale data.
If you specify a tag set (page 408), the client attempts to find a replica set member that matches the specified
tag set and directs reads to an arbitrary member from among the nearest group (page 408).
Read operations using the nearest (page 477) mode may return stale data.
Note: All operations read from a member of the nearest group of the replica set that matches the specified
read preference mode. The nearest (page 477) mode prefers low latency reads over a members primary or
secondary status.
For nearest (page 477), the client assembles a list of acceptable hosts based on tag set and then narrows that
list to the host with the shortest ping time and all other members of the set that are within the local threshold,
or acceptable latency. See Member Selection (page 408) for more information.

Use Cases

Depending on the requirements of an application, you can configure different applications to use different read preferences, or use different read preferences for different queries in the same application. Consider the following applications for different read preference strategies.
Maximize Consistency To avoid stale reads under all circumstances, use primary (page 476). This prevents all
queries when the set has no primary, which happens during elections, or when a majority of the replica set is not
accessible.
Maximize Availability To permit read operations when possible, Use primaryPreferred (page 476). When
theres a primary you will get consistent reads, but if there is no primary you can still query secondaries.
14 If your set has more than one secondary, and you use the secondary (page 476) read preference mode, consider the following effect. If
you have a three member replica set (page 392) with a primary and two secondaries, and if one secondary becomes unavailable, all secondary
(page 476) queries must target the remaining secondary. This will double the load on this secondary. Plan and provide capacity to support this as
needed.

8.4. Replication Reference

477

MongoDB Documentation, Release 2.4.9

Minimize Latency To always read from a low-latency node, use nearest (page 477). The driver or mongos will
read from the nearest member and those no more than 15 milliseconds 15 further away than the nearest member.
nearest (page 477) does not guarantee consistency. If the nearest member to your application server is a secondary
with some replication lag, queries could return stale data. nearest (page 477) only reflects network distance and
does not reflect I/O or CPU load.
Query From Geographically Distributed Members If the members of a replica set are geographically distributed,
you can create replica tags based that reflect the location of the instance and then configure your application to query
the members nearby.
For example, if members in east and west data centers are tagged (page 450) {dc: east} and {dc:
west}, your application servers in the east data center can read from nearby members with the following read
preference:
db.collection.find().readPref( { mode: 'nearest',
tags: [ {'dc': 'east'} ] } )

Although nearest (page 477) already favors members with low network latency, including the tag makes the choice
more predictable.
Reduce load on the primary To shift read load from the primary, use mode secondary (page 476). Although
secondaryPreferred (page 477) is tempting for this use case, it carries some risk: if all secondaries are unavailable and your set has enough arbiters to prevent the primary from stepping down, then the primary will receive all
traffic from clients. If the primary is unable to handle this load, queries will compete with writes. For this reason, use
secondary (page 476) to distribute read load to replica sets, not secondaryPreferred (page 477).
Read Preferences for Database Commands

Because some database commands read and return data from the database, all of the official drivers support full read
preference mode semantics (page 476) for the following commands:
group
mapReduce 16
aggregate
collStats
dbStats
count
distinct
geoNear
geoSearch
geoWalk
New in version 2.4: mongos adds support for routing commands to shards using read preferences. Previously
mongos sent all commands to shards primaries.

15

This threshold is configurable. See localThreshold for mongos or your driver documentation for the appropriate setting.
Only inline mapReduce operations that do not write data support read preference, otherwise these operations must run on the primary
members.
16

478

Chapter 8. Replication

CHAPTER 9

Sharding

Sharding is the process of storing data records across multiple machines and is MongoDBs approach to meeting the
demands of data growth. As the size of the data increases, a single machine may not be sufficient to store the data nor
provide an acceptable read and write throughput. Sharding solves the problem with horizontal scaling. With sharding,
you add more machines to support data growth and the demands of read and write operations.
Sharding Introduction (page 479) A high-level introduction to horizontal scaling, data partitioning, and sharded
clusters in MongoDB.
Sharding Concepts (page 484) The core documentation of sharded cluster features, configuration, architecture and
behavior.
Sharded Cluster Components (page 485) A sharded cluster consists of shards, config servers, and mongos
instances.
Sharded Cluster Architectures (page 489) Outlines the requirements for sharded clusters, and provides examples of several possible architectures for sharded clusters.
Sharded Cluster Behavior (page 490) Discusses the operations of sharded clusters with regards to the automatic balancing of data in a cluster and other related availability and security considerations.
Sharding Mechanics (page 500) Discusses the internal operation and behavior of sharded clusters, including
chunk migration, balancing, and the cluster metadata.
Sharded Cluster Tutorials (page 506) Tutorials that describe common procedures and administrative operations relevant to the use and maintenance of sharded clusters.
Sharding Reference (page 546) Reference for sharding-related functions and operations.

9.1 Sharding Introduction

Sharding is a method for storing data across multiple machines. MongoDB uses sharding to support deployments with
very large data sets and high throughput operations.

9.1.1 Purpose of Sharding

Database systems with large data sets and high throughput applications can challenge the capacity of a single server.
High query rates can exhaust the CPU capacity of the server. Larger data sets exceed the storage capacity of a single
machine. Finally, working set sizes larger than the systems RAM stress the I/O capacity of disk drives.
To address these issues of scales, database systems have two basic approaches: vertical scaling and sharding.

479

MongoDB Documentation, Release 2.4.9

Vertical scaling adds more CPU and storage resources to increase capacity. Scaling by adding capacity has limitations: high performance systems with large numbers of CPUs and large amount of RAM are disproportionately
more expensive than smaller systems. Additionally, cloud-based providers may only allow users to provision smaller
instances. As a result there is a practical maximum capability for vertical scaling.
Sharding, or horizontal scaling, by contrast, divides the data set and distributes the data over multiple servers, or
shards. Each shard is an independent database, and collectively, the shards make up a single logical database.

Figure 9.1: Diagram of a large collection with data distributed across 4 shards.
Sharding addresses the challenge of scaling to support high throughput and large data sets:
Sharding reduces the number of operations each shard handles. Each shard processes fewer operations as the
cluster grows. As a result, shared clusters can increase capacity and throughput horizontally.
For example, to insert data, the application only needs to access the shard responsible for that records.
Sharding reduces the amount of data that each server needs to store. Each shard stores less data as the cluster
grows.
For example, if a database has a 1 terabyte data set, and there are 4 shards, then each shard might hold only
256GB of data. If there are 40 shards, then each shard might hold only 25GB of data.

480

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

9.1.2 Sharding in MongoDB

MongoDB supports sharding through the configuration of a sharded clusters.

Figure 9.2: Diagram of a sample sharded cluster for production purposes. Contains exactly 3 config servers, 2 or more
mongos query routers, and at least 2 shards. The shards are replica sets.
Sharded cluster has the following components: shards, query routers and config servers.
Shards store the data. To provide high availability and data consistency, in a production sharded cluster, each shard is
a replica set 1 . For more information on replica sets, see Replica Sets (page 381).
Query Routers, or mongos instances, interface with client applications and direct operations to the appropriate shard
or shards. The query router processes and targets operations to shards and then returns results to the clients. A sharded
cluster can contain more than one query router to divide the client request load. A client sends requests to one query
router. Most sharded cluster have many query routers.
Config servers store the clusters metadata. This data contains a mapping of the clusters data set to the shards. The
query router uses this metadata to target operations to specific shards. Production sharded clusters have exactly 3
config servers.

9.1.3 Data Partitioning

MongoDB distributes data, or shards, at the collection level. Sharding partitions a collections data by the shard key.
1 For development and testing purposes only, each shard can be a single mongod instead of a replica set. Do not deploy production clusters
without 3 config servers.

9.1. Sharding Introduction

481

MongoDB Documentation, Release 2.4.9

Shard Keys
To shard a collection, you need to select a shard key. A shard key is either an indexed field or an indexed compound
field that exists in every document in the collection. MongoDB divides the shard key values into chunks and distributes
the chunks evenly across the shards. To divide the shard key values into chunks, MongoDB uses either range based
partitioning and hash based partitioning. See Shard Keys (page 492) for more information.
Range Based Sharding
For range-based sharding, MongoDB divides the data set into ranges determined by the shard key values to provide
range based partitioning. Consider a numeric shard key: If you visualize a number line that goes from negative
infinity to positive infinity, each value of the shard key falls at some point on that line. MongoDB partitions this line
into smaller, non-overlapping ranges called chunks where a chunk is range of values from some minimum value to
some maximum value.
Given a range based partitioning system, documents with close shard key values are likely to be in the same chunk,
and therefore on the same shard.

Figure 9.3: Diagram of the shard key value space segmented into smaller ranges or chunks.

Hash Based Sharding

For hash based partitioning, MongoDB computes a hash of a fields value, and then uses these hashes to create chunks.
With hash based partitioning, two documents with close shard key values are unlikely to be part of the same chunk.
This ensures a more random distribution of a collection in the cluster.
Performance Distinctions between Range and Hash Based Partitioning
Range based partitioning supports more efficient range queries. Given a range query on the shard key, the query router
can easily determine which chunks overlap that range and route the query to only those shards that contain these
chunks.
However, range based partitioning can result in an uneven distribution of data, which may negate some of the benefits
of sharding. For example, if the shard key is a linearly increasing field, such as time, then all requests for a given time
range will map to the same chunk, and thus the same shard. In this situation, a small set of shards may receive the
majority of requests and the system would not scale very well.

482

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Figure 9.4: Diagram of the hashed based segmentation.

Hash based partitioning, by contrast, ensures an even distribution of data at the expense of efficient range queries.
Hashed key values results in random distribution of data across chunks and therefore shards. But random distribution
makes it more likely that a range query on the shard key will not be able to target a few shards but would more likely
query every shard in order to return a result.

9.1.4 Maintaining a Balanced Data Distribution

The addition of new data or the addition of new servers can result in data distribution imbalances within the cluster,
such as a particular shard contains significantly more chunks than another shard or a size of a chunk is significantly
greater than other chunk sizes.
MongoDB ensures a balanced cluster using two background process: splitting and the balancer.
Splitting
Splitting is a background process that keeps chunks from growing too large. When a chunk grows beyond a specified
chunk size (page 504), MongoDB splits the chunk in half. Inserts and updates triggers splits. Splits are a efficient
meta-data change. To create splits, MongoDB does not migrate any data or affect the shards.

Figure 9.5: Diagram of a shard with a chunk that exceeds the default chunk size of 64 MB and triggers a split of the
chunk into two chunks.

9.1. Sharding Introduction

483

MongoDB Documentation, Release 2.4.9

Balancing
The balancer (page 501) is a background process that manages chunk migrations. The balancer runs in all of the query
routers in a cluster.
When the distribution of a sharded collection in a cluster is uneven, the balancer process migrates chunks from the
shard that has the largest number of chunks to the shard with the least number of chunks until the collection balances.
For example: if collection users has 100 chunks on shard 1 and 50 chunks on shard 2, the balancer will migrate
chunks from shard 1 to shard 2 until the collections achieves balance.
The shards manage chunk migrations as a background operation. During migration, all requests for a chunks data
address the origin shard.
In a chunk migration, the destination shard receives all the documents in the chunk from the origin shard. Then, the
destination shard captures and applies all changes made to the data during migration process. Finally, the destination
shard updates the metadata regarding the location of the on config server.
If theres an error during the migration, the balancer aborts the process leaving the chunk on the origin shard. MongoDB removes the chunks data from the origin shard after the migration completes successfully.

Figure 9.6: Diagram of a collection distributed across three shards. For this collection, the difference in the number of
chunks between the shards reaches the migration thresholds (in this case, 2) and triggers migration.

Adding and Removing Shards from the Cluster

Adding a shard to a cluster creates an imbalance since the new shard has no chunks. While MongoDB begins migrating
data to the new shard immediately, it can take some time before the cluster balances.
When removing a shard, the balancer migrates all chunks from a shard to other shards. After migrating all data and
updating the meta data, you can safely remove the shard.

9.2 Sharding Concepts

These documents present the details of sharding in MongoDB. These include the components, the architectures, and the
behaviors of MongoDB sharded clusters. For an overview of sharding and sharded clusters, see Sharding Introduction
(page 479).
Sharded Cluster Components (page 485) A sharded cluster consists of shards, config servers, and mongos instances.
Shards (page 485) A shard is a mongod instance that holds a part of the sharded collections data.

484

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Config Servers (page 488) Config servers hold the metadata about the cluster, such as the shard location of the
data.
Sharded Cluster Architectures (page 489) Outlines the requirements for sharded clusters, and provides examples of
several possible architectures for sharded clusters.
Sharded Cluster Requirements (page 489) Discusses the requirements for sharded clusters in MongoDB.
Production Cluster Architecture (page 490) Sharded cluster for production has component requirements to
provide redundancy and high availability.
Sharded Cluster Behavior (page 490) Discusses the operations of sharded clusters with regards to the automatic balancing of data in a cluster and other related availability and security considerations.
Shard Keys (page 492) MongoDB uses the shard key to divide a collections data across the clusters shards.
Sharded Cluster High Availability (page 494) Sharded clusters provide ways to address some availability concerns.
Sharded Cluster Query Routing (page 496) The clusters routers, or mongos instances, send reads and writes
to the relevant shard or shards.
Sharding Mechanics (page 500) Discusses the internal operation and behavior of sharded clusters, including chunk
migration, balancing, and the cluster metadata.
Sharded Collection Balancing (page 501) Balancing distributes a sharded collections data cluster to all of the
shards.
Sharded Cluster Metadata (page 505) The cluster maintains internal metadata that reflects the location of data
within the cluster.

9.2.1 Sharded Cluster Components

Sharded clusters implement sharding. A sharded cluster consists of the following components:
Shards A shard is a MongoDB instance that holds a subset of a collections data. Each shard is either a single
mongod instance or a replica set. In production, all shards are replica sets. For more information see Shards
(page 485).
Config Servers Each config server (page 488) is a mongod instance that holds metadata about the cluster. The
metadata maps chunks to shards. For more information, see Config Servers (page 488).
Routing Instances Each router is a mongos instance that routes the reads and writes from applications to the shards.
Applications do not access the shards directly. For more information see Sharded Cluster Query Routing
(page 496).
Enable sharding in MongoDB on a per-collection basis. For each collection you shard, you will specify a shard key
for that collection.
Deploy a sharded cluster, see Deploy a Sharded Cluster (page 507).
Shards
A shard is a replica set or a single mongod that contains a subset of the data for the sharded cluster. Together, the
clusters shards hold the entire data set for the cluster.
Typically each shard is a replica set. The replica set provides redundancy and high availability for the data in each
shard.
Important: MongoDB shards data on a per collection basis. You must access all data in a sharded cluster via the
mongos instances. If you connect directly to a shard, you will see only its fraction of the clusters data. There is no
9.2. Sharding Concepts

485

MongoDB Documentation, Release 2.4.9

Figure 9.7: Diagram of a sharded cluster.

486

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

particular order to the data set on a specific shard. MongoDB does not guarantee that any two contiguous chunks will
reside on a single shard.

Primary Shard

Every database has a primary 2 shard that holds all the un-sharded collections in that database.

Figure 9.8: Diagram of a primary shard. A primary shard contains non-sharded collections as well as chunks of
documents from sharded collections. Shard A is the primary shard.
To change the primary shard for a database, use the movePrimary command.
Warning: The movePrimary command can be expensive because it copies all non-sharded data to the new
shard. During this time, this data will be unavailable for other operations.
When you deploy a new sharded cluster, the first shard becomes the primary shard for all existing databases before
enabling sharding. Databases created subsequently may reside on any shard in the cluster.
Shard Status

Use the sh.status() method in the mongo shell to see an overview of the cluster. This reports includes which
shard is primary for the database and the chunk distribution across the shards. See sh.status() method for more
details.
2

The term primary shard has nothing to do with the term primary in the context of replica sets.

9.2. Sharding Concepts

487

MongoDB Documentation, Release 2.4.9

Config Servers
Config servers are special mongod instances that store the metadata (page 505) for a sharded cluster. Config servers
use a two-phase commit to ensure immediate consistency and reliability. Config servers do not run as replica sets. All
config servers must be available to deploy a sharded cluster or to make any changes to cluster metadata.
A production sharded cluster has exactly three config servers. For testing purposes you may deploy a cluster with a
single config server. But to ensure redundancy and safety in production, you should always use three.
Warning: If your cluster has a single config server, then the config server is a single point of failure. If the config
server is inaccessible, the cluster is not accessible. If you cannot recover the data on a config server, the cluster
will be inoperable.
Always use three config servers for production deployments.
Config servers store metadata for a single sharded cluster. Each cluster must have its own config servers.
Tip
Use CNAMEs to identify your config servers to the cluster so that you can rename and renumber your config servers
without downtime.

Config Database

Config servers store the metadata in the config database (page 547). The mongos instances cache this data and use it
to route reads and writes to shards.
Read and Write Operations on Config Servers

MongoDB only writes data to the config server in the following cases:
To create splits in existing chunks. For more information, see chunk splitting (page 504).
To migrate a chunk between shards. For more information, see chunk migration (page 502).
MongoDB reads data from the config server data in the following cases:
A new mongos starts for the first time, or an existing mongos restarts.
After a chunk migration, the mongos instances update themselves with the new cluster metadata.
MongoDB also uses the config server to manage distributed locks.
Config Server Availability

If one or two config servers become unavailable, the clusters metadata becomes read only. You can still read and
write data from the shards, but no chunk migrations or splits will occur until all three servers are available.
If all three config servers are unavailable, you can still use the cluster if you do not restart the mongos instances
until after the config servers are accessible again. If you restart the mongos instances before the config servers are
available, the mongos will be unable to route reads and writes.
Clusters become inoperable without the cluster metadata. Always, ensure that the config servers remain available and
intact. As such, backups of config servers are critical. The data on the config server is small compared to the data
stored in a cluster. This means the config server has a relatively low activity load, and the config server does not need
to be always available to support a sharded cluster. As a result, it is easy to back up the config servers.

488

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

If the name or address that a sharded cluster uses to connect to a config server changes, you must restart every mongod
and mongos instance in the sharded cluster. Avoid downtime by using CNAMEs to identify config servers within the
MongoDB deployment.
See Renaming Config Servers and Cluster Availability (page 495) for more information.

9.2.2 Sharded Cluster Architectures

The following documents introduce deployment patterns for sharded clusters.
Sharded Cluster Requirements (page 489) Discusses the requirements for sharded clusters in MongoDB.
Production Cluster Architecture (page 490) Sharded cluster for production has component requirements to provide
redundancy and high availability.
Sharded Cluster Test Architecture (page 490) Sharded clusters for testing and development can have fewer components.
Sharded Cluster Requirements
While sharding is a powerful and compelling feature, sharded clusters have significant infrastructure requirements
and increases the overall complexity of a deployment. As a result, only deploy sharded clusters when indicated by
application and operational requirements
Sharding is the only solution for some classes of deployments. Use sharded clusters if:
your data set approaches or exceeds the storage capacity of a single MongoDB instance.
the size of your systems active working set will soon exceed the capacity of your systems maximum RAM.
a single MongoDB instance cannot meet the demands of your write operations, and all other approaches have
not reduced contention.
If these attributes are not present in your system, sharding will only add complexity to your system without adding
much benefit.
Important: It takes time and resources to deploy sharding. If your system has already reached or exceeded its
capacity, it will be difficult to deploy sharding without impacting your application.
As a result, if you think you will need to partition your database in the future, do not wait until your system is
overcapacity to enable sharding.
When designing your data model, take into consideration your sharding needs.
Data Quantity Requirements

Your cluster should manage a large quantity of data if sharding is to have an effect. The default chunk size is 64
megabytes. And the balancer (page 501) will not begin moving data across shards until the imbalance of chunks among
the shards exceeds the migration threshold (page 502). In practical terms, unless your cluster has many hundreds of
megabytes of data, your data will remain on a single shard.
In some situations, you may need to shard a small collection of data. But most of the time, sharding a small collection
is not worth the added complexity and overhead unless you need additional write capacity. If you have a small data
set, a properly configured single MongoDB instance or a replica set will usually be enough for your persistence layer
needs.

9.2. Sharding Concepts

489

MongoDB Documentation, Release 2.4.9

Chunk size is user configurable. For most deployments, the default value is of 64 megabytes is ideal. See
Chunk Size (page 504) for more information.
Production Cluster Architecture
In a production cluster, you must ensure that data is redundant and that your systems are highly available. To that end,
a production cluster must have the following components:
Components

Config Servers Three config servers (page 488). Each config server must be on separate machines. A single sharded
cluster must have exclusive use of its config servers (page 488). If you have multiple sharded clusters, you will need
to have a group of config servers for each cluster.
Shards Two or more replica sets. These replica sets are the shards. For information on replica sets, see Replication
(page 377).
Query Routers (mongos) One or more mongos instances. The mongos instances are the routers for the cluster.
Typically, deployments have one mongos instance on each application server.
You may also deploy a group of mongos instances and use a proxy/load balancer between the application and the
mongos. In these deployments, you must configure the load balancer for client affinity so that every connection from
a single client reaches the same mongos.
Because cursors and other resources are specific to an single mongos instance, each client must interact with only
one mongos instance.
Example

Sharded Cluster Test Architecture

Warning: Use the test cluster architecture for testing and development only.
For testing and development, you can deploy a minimal sharded clusters cluster. These non-production clusters have
the following components:
One config server (page 488).
At least one shard. Shards are either replica sets or a standalone mongod instances.
One mongos instance.
See
Production Cluster Architecture (page 490)

9.2.3 Sharded Cluster Behavior

These documents address the distribution of data and queries to a sharded cluster as well as specific security and
availability considerations for sharded clusters.

490

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Figure 9.9: Diagram of a sample sharded cluster for production purposes. Contains exactly 3 config servers, 2 or more
mongos query routers, and at least 2 shards. The shards are replica sets.

9.2. Sharding Concepts

491

MongoDB Documentation, Release 2.4.9

Figure 9.10: Diagram of a sample sharded cluster for testing/development purposes only. Contains only 1 config
server, 1 mongos router, and at least 1 shard. The shard can be either a replica set or a standalone mongod instance.
Shard Keys (page 492) MongoDB uses the shard key to divide a collections data across the clusters shards.
Sharded Cluster High Availability (page 494) Sharded clusters provide ways to address some availability concerns.
Sharded Cluster Query Routing (page 496) The clusters routers, or mongos instances, send reads and writes to the
relevant shard or shards.
Shard Keys
The shard key determines the distribution of the collections documents among the clusters shards. The shard key is
either an indexed field or an indexed compound field that exists in every document in the collection.
MongoDB partitions data in the collection using ranges of shard key values. Each range, or chunk, defines a nonoverlapping range of shard key values. MongoDB distributes the chunks, and their documents, among the shards in
the cluster.
When a chunk grows beyond the chunk size (page 504), MongoDB splits the chunk into smaller chunks, always based
on ranges in the shard key.
Considerations

Shard keys are immutable and cannot be changed after insertion. See the system limits for sharded cluster for more
information.
The index on the shard key cannot be a multikey index (page 324).

492

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Figure 9.11: Diagram of the shard key value space segmented into smaller ranges or chunks.
Hashed Shard Keys

New in version 2.4.

Hashed shard keys use a hashed index (page 343) of a single field as the shard key to partition data across your sharded
cluster.
The field you choose as your hashed shard key should have a good cardinality, or large number of different values.
Hashed keys work well with fields that increase monotonically like ObjectId values or timestamps.
If you shard an empty collection using a hashed shard key, MongoDB will automatically create and migrate
chunks so that each shard has two chunks. You can control how many chunks MongoDB will create with the
numInitialChunks parameter to shardCollection or by manually creating chunks on the empty collection
using the split command.
To shard a collection using a hashed shard key, see Shard a Collection Using a Hashed Shard Key (page 513).
Tip
MongoDB automatically computes the hashes when resolving queries using hashed indexes. Applications do not need
to compute hashes.

Impacts of Shard Keys on Cluster Operations

The shard key affects write and query performance by determining how the MongoDB partitions data in the cluster
and how effectively the mongos instances can direct operations to the cluster. Consider the following operational
impacts of shard key selection:
Write Scaling Some possible shard keys will allow your application to take advantage of the increased write capacity
that the cluster can provide, while others do not. Consider the following example where you shard by the values of the
default _id field, which is ObjectId.
MongoDB generates ObjectId values upon document creation to produce a unique identifier for the object. However, the most significant bits of data in this value represent a time stamp, which means that they increment in a regular
and predictable pattern. Even though this value has high cardinality (page 512), when using this, any date, or other
monotonically increasing number as the shard key, all insert operations will be storing data into a single chunk, and
therefore, a single shard. As a result, the write capacity of this shard will define the effective write capacity of the
cluster.
9.2. Sharding Concepts

493

MongoDB Documentation, Release 2.4.9

A shard key that increases monotonically will not hinder performance if you have a very low insert rate, or if most
of your write operations are update() operations distributed through your entire data set. Generally, choose shard
keys that have both high cardinality and will distribute write operations across the entire cluster.
Typically, a computed shard key that has some amount of randomness, such as ones that include a cryptographic
hash (i.e. MD5 or SHA1) of other content in the document, will allow the cluster to scale write operations. However,
random shard keys do not typically provide query isolation (page 494), which is another important characteristic of
shard keys.
New in version 2.4: MongoDB makes it possible to shard a collection on a hashed index. This can greatly improve
write scaling. See Shard a Collection Using a Hashed Shard Key (page 513).
Querying The mongos provides an interface for applications to interact with sharded clusters that hides the complexity of data partitioning. A mongos receives queries from applications, and uses metadata from the config server
(page 488), to route queries to the mongod instances with the appropriate data. While the mongos succeeds in making all querying operational in sharded environments, the shard key you select can have a profound affect on query
performance.
See also:
The Sharded Cluster Query Routing (page 496) and config server (page 488) sections for a more general overview of
querying in sharded environments.
Query Isolation The fastest queries in a sharded environment are those that mongos will route to a single shard,
using the shard key and the cluster meta data from the config server (page 488). For queries that dont include the
shard key, mongos must query all shards, wait for their response and then return the result to the application. These
scatter/gather queries can be long running operations.
If your query includes the first component of a compound shard key 3 , the mongos can route the query directly to a
single shard, or a small number of shards, which provides better performance. Even if you query values of the shard
key reside in different chunks, the mongos will route queries directly to specific shards.
To select a shard key for a collection:
determine the most commonly included fields in queries for a given application
find which of these operations are most performance dependent.
If this field has low cardinality (i.e not sufficiently selective) you should add a second field to the shard key making a
compound shard key. The data may become more splittable with a compound shard key.
See
Sharded Cluster Query Routing (page 496) for more information on query operations in the context of sharded clusters.

Sorting In sharded systems, the mongos performs a merge-sort of all sorted query results from the shards. See
Sharded Cluster Query Routing (page 496) and Use Indexes to Sort Query Results (page 371) for more information.
Sharded Cluster High Availability
A production (page 490) cluster has no single point of failure. This section introduces the availability concerns for
MongoDB deployments in general and highlights potential failure scenarios and available resolutions.
3 In many ways, you can think of the shard key a cluster-wide unique index. However, be aware that sharded systems cannot enforce clusterwide unique indexes unless the unique field is in the shard key. Consider the Index Concepts (page 318) page for more information on indexes and
compound indexes.

494

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Application Servers or mongos Instances Become Unavailable

If each application server has its own mongos instance, other application servers can continue access the database.
Furthermore, mongos instances do not maintain persistent state, and they can restart and become unavailable without
losing any state or data. When a mongos instance starts, it retrieves a copy of the config database and can begin
routing queries.
A Single mongod Becomes Unavailable in a Shard

Replica sets (page 377) provide high availability for shards. If the unavailable mongod is a primary, then the replica
set will elect (page 397) a new primary. If the unavailable mongod is a secondary, and it disconnects the primary and
secondary will continue to hold all data. In a three member replica set, even if a single member of the set experiences
catastrophic failure, two other members have full copies of the data. 4
Always investigate availability interruptions and failures. If a system is unrecoverable, replace it and create a new
member of the replica set as soon as possible to replace the lost redundancy.
All Members of a Replica Set Become Unavailable

If all members of a replica set within a shard are unavailable, all data held in that shard is unavailable. However, the
data on all other shards will remain available, and its possible to read and write data to the other shards. However,
your application must be able to deal with partial results, and you should investigate the cause of the interruption and
attempt to recover the shard as soon as possible.
One or Two Config Databases Become Unavailable

Three distinct mongod instances provide the config database using a special two-phase commits to maintain consistent
state between these mongod instances. Cluster operation will continue as normal but chunk migration (page 501) and
the cluster can create no new chunk splits (page 537). Replace the config server as soon as possible. If all config
databases become unavailable, the cluster can become inoperable.
Note: All config servers must be running and available when you first initiate a sharded cluster.

Renaming Config Servers and Cluster Availability

If the name or address that a sharded cluster uses to connect to a config server changes, you must restart every mongod
and mongos instance in the sharded cluster. Avoid downtime by using CNAMEs to identify config servers within the
MongoDB deployment.
To avoid downtime when renaming config servers, use DNS names unrelated to physical or virtual hostnames to refer
to your config servers (page 488).
Generally, refer to each config server using the DNS alias (e.g. a CNAME record). When specifying the config server
connection string to mongos, use these names. These records make it possible to change the IP address or rename
config servers without changing the connection string and without having to restart the entire cluster.
4 If an unavailable secondary becomes available while it still has current oplog entries, it can catch up to the latest state of the set using the
normal replication process, otherwise it must perform an initial sync.

9.2. Sharding Concepts

495

MongoDB Documentation, Release 2.4.9

Shard Keys and Cluster Availability

The most important consideration when choosing a shard key are:

to ensure that MongoDB will be able to distribute data evenly among shards, and
to scale writes across the cluster, and
to ensure that mongos can isolate most queries to a specific mongod.
Furthermore:
Each shard should be a replica set, if a specific mongod instance fails, the replica set members will elect another
to be primary and continue operation. However, if an entire shard is unreachable or fails for some reason, that
data will be unavailable.
If the shard key allows the mongos to isolate most operations to a single shard, then the failure of a single shard
will only render some data unavailable.
If your shard key distributes data required for every operation throughout the cluster, then the failure of the entire
shard will render the entire cluster unavailable.
In essence, this concern for reliability simply underscores the importance of choosing a shard key that isolates query
operations to a single shard.
Sharded Cluster Query Routing
MongoDB mongos instances route queries and write operations to shards in a sharded cluster. mongos provide the
only interface to a sharded cluster from the perspective of applications. Applications never connect or communicate
directly with the shards.
The mongos tracks what data is on which shard by caching the metadata from the config servers (page 488). The
mongos uses the metadata to route operations from applications and clients to the mongod instances. A mongos
has no persistent state and consumes minimal system resources.
The most common practice is to run mongos instances on the same systems as your application servers, but you can
maintain mongos instances on the shards or on other dedicated resources.
Note: Changed in version 2.1.
Some aggregation operations using the aggregate command (i.e. db.collection.aggregate()) will cause
mongos instances to require more CPU resources than in previous versions. This modified performance profile may
dictate alternate architecture decisions if you use the aggregation framework extensively in a sharded environment.

Routing Process

A mongos instance uses the following processes to route queries and return results.
How mongos Determines which Shards Receive a Query A mongos instance routes a query to a cluster by:
1. Determining the list of shards that must receive the query.
2. Establishing a cursor on all targeted shards.
In some cases, when the shard key or a prefix of the shard key is a part of the query, the mongos can route the query
to a subset of the shards. Otherwise, the mongos must direct the query to all shards that hold documents for that
collection.

496

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Example
Given the following shard key:
{ zipcode: 1, u_id: 1, c_date: 1 }

Depending on the distribution of chunks in the cluster, the mongos may be able to target the query at a subset of
shards, if the query contains the following fields:
{ zipcode: 1 }
{ zipcode: 1, u_id: 1 }
{ zipcode: 1, u_id: 1, c_date: 1 }

How mongos Handles Query Modifiers If the result of the query is not sorted, the mongos instance opens a result
cursor that round robins results from all cursors on the shards.
Changed in version 2.0.5: In versions prior to 2.0.5, the mongos exhausted each cursor, one by one.
If the query specifies sorted results using the sort() cursor method, the mongos instance passes the $orderby
option to the shards. When the mongos receives results it performs an incremental merge sort of the results while
returning them to the client.
If the query limits the size of the result set using the limit() cursor method, the mongos instance passes that limit
to the shards and then re-applies the limit to the result before returning the result to the client.
If the query specifies a number of records to skip using the skip() cursor method, the mongos cannot pass the skip
to the shards, but rather retrieves unskipped results from the shards and skips the appropriate number of documents
when assembling the complete result. However, when used in conjunction with a limit(), the mongos will pass
the limit plus the value of the skip() to the shards to improve the efficiency of these operations.
Detect Connections to mongos Instances

To detect if the MongoDB instance that your client is connected to is mongos, use the isMaster command. When
a client connects to a mongos, isMaster returns a document with a msg field that holds the string isdbgrid. For
example:
{
"ismaster" : true,
"msg" : "isdbgrid",
"maxBsonObjectSize" : 16777216,
"ok" : 1
}

If the application is instead connected to a mongod, the returned document does not include the isdbgrid string.
Broadcast Operations and Targeted Operations

In general, operations in a sharded environment are either:

Broadcast to all shards in the cluster that hold documents in a collection
Targeted at a single shard or a limited group of shards, based on the shard key
For best performance, use targeted operations whenever possible. While some operations must broadcast to all shards,
you can ensure MongoDB uses targeted operations whenever possible by always including the shard key.

9.2. Sharding Concepts

497

MongoDB Documentation, Release 2.4.9

Broadcast Operations mongos instances broadcast queries to all shards for the collection unless the mongos can
determine which shard or subset of shards stores this data.

Figure 9.12: Read operations to a sharded cluster. Query criteria does not include the shard key. The query router
mongos must broadcast query to all shards for the collection.
Multi-update operations are always broadcast operations.
The remove() operation is always a broadcast operation, unless the operation specifies the shard key in full.
Targeted Operations All insert() operations target to one shard.
All single update() (including upsert operations) and remove() operations must target to one shard.
Important: All single update() and remove() operations must include the shard key or the _id field in
the query specification. update() or remove() operations that affect a single document in a sharded collection
without the shard key or the _id field return an error.
For queries that include the shard key or portion of the shard key, mongos can target the query at a specific shard or
set of shards. This is the case only if the portion of the shard key included in the query is a prefix of the shard key. For
example, if the shard key is:

498

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

{ a: 1, b: 1, c: 1 }

The mongos program can route queries that include the full shard key or either of the following shard key prefixes at
a specific shard or set of shards:
{ a: 1 }
{ a: 1, b: 1 }

Figure 9.13: Read operations to a sharded cluster. Query criteria includes the shard key. The query router mongos
can target the query to the appropriate shard or shards.
Depending on the distribution of data in the cluster and the selectivity of the query, mongos may still have to contact
multiple shards 5 to fulfill these queries.
Sharded and Non-Sharded Data

Sharding operates on the collection level. You can shard multiple collections within a database or have multiple
databases with sharding enabled. 6 However, in production deployments, some databases and collections will use
sharding, while other databases and collections will only reside on a single shard.
5

mongos will route some queries, even some that include the shard key, to all shards, if needed.
As you configure sharding, you will use the enableSharding command to enable sharding for a database. This simply makes it possible
to use the shardCollection command on a collection within that database.
6

9.2. Sharding Concepts

499

MongoDB Documentation, Release 2.4.9

Figure 9.14: Diagram of a primary shard. A primary shard contains non-sharded collections as well as chunks of
documents from sharded collections. Shard A is the primary shard.
Regardless of the data architecture of your sharded cluster, ensure that all queries and operations use the mongos
router to access the data cluster. Use the mongos even for operations that do not impact the sharded data.
Related
Sharded Cluster Security (page 241)

9.2.4 Sharding Mechanics

The following documents describe sharded cluster processes.
Sharded Collection Balancing (page 501) Balancing distributes a sharded collections data cluster to all of the
shards.
Chunk Migration Across Shards (page 502) MongoDB migrates chunks to shards as part of the balancing process.
Chunk Splits in a Sharded Cluster (page 504) When a chunk grows beyond the configured size, MongoDB splits the
chunk in half.
Shard Key Indexes (page 505) Sharded collections must keep an index that starts with the shard key.
Sharded Cluster Metadata (page 505) The cluster maintains internal metadata that reflects the location of data within
the cluster.

500

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Figure 9.15: Diagram of applications/drivers issuing queries to mongos for unsharded collection as well as sharded
collection. Config servers not shown.
Sharded Collection Balancing
Balancing is the process MongoDB uses to distribute data of a sharded collection evenly across a sharded cluster.
When a shard has too many of a sharded collections chunks compared to other shards, MongoDB automatically
balances the the chunks across the shards. The balancing procedure for sharded clusters is entirely transparent to the
user and application layer.
Cluster Balancer

The balancer process is responsible for redistributing the chunks of a sharded collection evenly among the shards for
every sharded collection. By default, the balancer process is always enabled.
Any mongos instance in the cluster can start a balancing round. When a balancer process is active, the responsible
mongos acquires a lock by modifying a document in the lock collection in the Config Database (page 547).
Note: Changed in version 2.0: Before MongoDB version 2.0, large differences in timekeeping (i.e. clock skew)
between mongos instances could lead to failed distributed locks. This carries the possibility of data loss, particularly
with skews larger than 5 minutes. Always use the network time protocol (NTP) by running ntpd on your servers to
minimize clock skew.
To address uneven chunk distribution for a sharded collection, the balancer migrates chunks (page 502) from shards
with more chunks to shards with a fewer number of chunks. The balancer migrates the chunks, one at a time, until
there is an even dispersion of chunks for the collection across the shards.
Chunk migrations carry some overhead in terms of bandwidth and workload, both of which can impact database
performance. The balancer attempts to minimize the impact by:
Moving only one chunk at a time. See also Chunk Migration Queuing (page 503).

9.2. Sharding Concepts

501

MongoDB Documentation, Release 2.4.9

Starting a balancing round only when the difference in the number of chunks between the shard with the greatest
number of chunks for a sharded collection and the shard with the lowest number of chunks for that collection
reaches the migration threshold (page 502).
You may disable the balancer temporarily for maintenance. See Disable the Balancer (page 533) for details.
You can also limit the window during which the balancer runs to prevent it from impacting production traffic. See
Schedule the Balancing Window (page 532) for details.
Note: The specification of the balancing window is relative to the local time zone of all individual mongos instances
in the cluster.
See also:
Manage Sharded Cluster Balancer (page 531).
Migration Thresholds

To minimize the impact of balancing on the cluster, the balancer will not begin balancing until the distribution of
chunks for a sharded collection has reached certain thresholds. The thresholds apply to the difference in number
of chunks between the shard with the most chunks for the collection and the shard with the fewest chunks for that
collection. The balancer has the following thresholds:
Changed in version 2.2: The following thresholds appear first in 2.2. Prior to this release, a balancing round would
only start if the shard with the most chunks had 8 more chunks than the shard with the least number of chunks.
Number of Chunks
Fewer than 20
21-80
Greater than 80

Migration Threshold
2
4
8

Once a balancing round starts, the balancer will not stop until, for the collection, the difference between the number
of chunks on any two shards for that collection is less than two or a chunk migration fails.
Shard Size

By default, MongoDB will attempt to fill all available disk space with data on every shard as the data set grows. To
ensure that the cluster always has the capacity to handle data growth, monitor disk usage as well as other performance
metrics.
When adding a shard, you may set a maximum size for that shard. This prevents the balancer from migrating chunks
to the shard when the value of mapped exceeds the maximum size. Use the maxSize parameter of the addShard
command to set the maximum size for the shard.
See also:
Change the Maximum Storage Size for a Given Shard (page 530) and Monitoring for MongoDB (page 138).
Chunk Migration Across Shards
Chunk migration moves the chunks of a sharded collection from one shard to another and is part of the balancer
(page 501) process.

502

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Figure 9.16: Diagram of a collection distributed across three shards. For this collection, the difference in the number
of chunks between the shards reaches the migration thresholds (in this case, 2) and triggers migration.
Chunk Migration

MongoDB migrates chunks in a sharded cluster to distribute the chunks of a sharded collection evenly among shards.
Migrations may be either:
Manual. Only use manual migration in limited cases, such as to distribute data during bulk inserts. See Migrating
Chunks Manually (page 538) for more details.
Automatic. The balancer (page 501) process automatically migrates chunks when there is an uneven distribution
of a sharded collections chunks across the shards. See Migration Thresholds (page 502) for more details.
All chunk migrations use the following procedure:
1. The balancer process sends the moveChunk command to the source shard.
2. The source starts the move with an internal moveChunk command. During the migration process, operations
to the chunk route to the source shard. The source shard is responsible for incoming write operations for the
chunk.
3. The destination shard begins requesting documents in the chunk and starts receiving copies of the data.
4. After receiving the final document in the chunk, the destination shard starts a synchronization process to ensure
that it has the changes to the migrated documents that occurred during the migration.
5. When fully synchronized, the destination shard connects to the config database and updates the cluster metadata
with the new location for the chunk.
6. After the destination shard completes the update of the metadata, and once there are no open cursors on the
chunk, the source shard deletes its copy of the documents.
Changed in version 2.4: If the balancer needs to perform additional chunk migrations from the source shard,
the balancer can start the next chunk migration without waiting for the current migration process to finish this
deletion step. See Chunk Migration Queuing (page 503).
The migration process ensures consistency and maximizes the availability of chunks during balancing.
Chunk Migration Queuing Changed in version 2.4.
To migrate multiple chunks from a shard, the balancer migrates the chunks one at a time. However, the balancer does
not wait for the current migrations delete phase to complete before starting the next chunk migration. See Chunk
Migration (page 503) for the chunk migration process and the delete phase.

9.2. Sharding Concepts

503

MongoDB Documentation, Release 2.4.9

This queuing behavior allows shards to unload chunks more quickly in cases of heavily imbalanced cluster, such as
when performing initial data loads without pre-splitting and when adding new shards.
This behavior also affect the moveChunk command, and migration scripts that use the moveChunk command may
proceed more quickly.
In some cases, the delete phases may persist longer. If multiple delete phases are queued but not yet complete, a crash
of the replica sets primary can orphan data from multiple migrations.
Chunk Migration Write Concern Changed in version 2.4: While copying and deleting data during migrations, the
balancer waits for replication to secondaries (page 49) for every document. This slows the potential speed of a chunk
migration but ensures that a large number of chunk migrations cannot affect the availability of a sharded cluster.
See also Secondary Throttle in the v2.2 Manual7 .
Chunk Splits in a Sharded Cluster
As chunks grow beyond the specified chunk size (page 504) a mongos instance will attempt to split the chunk in half.
Splits may lead to an uneven distribution of the chunks for a collection across the shards. In such cases, the mongos
instances will initiate a round of migrations to redistribute chunks across shards. See Sharded Collection Balancing
(page 501) for more details on balancing chunks across shards.

Figure 9.17: Diagram of a shard with a chunk that exceeds the default chunk size of 64 MB and triggers a split of the
chunk into two chunks.

Chunk Size

The default chunk size in MongoDB is 64 megabytes. You can increase or reduce the chunk size (page 539), mindful
of its effect on the clusters efficiency.
1. Small chunks lead to a more even distribution of data at the expense of more frequent migrations. This creates
expense at the query routing (mongos) layer.
2. Large chunks lead to fewer migrations. This is more efficient both from the networking perspective and in terms
of internal overhead at the query routing layer. But, these efficiencies come at the expense of a potentially more
uneven distribution of data.
For many deployments, it makes sense to avoid frequent and potentially spurious migrations at the expense of a slightly
less evenly distributed data set.
7 http://docs.mongodb.org/v2.2/tutorial/configure-sharded-cluster-balancer/#sharded-cluster-config-secondary-throttle

504

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Limitations

Changing the chunk size affects when chunks split but there are some limitations to its effects.
Automatic splitting only occurs during inserts or updates. If you lower the chunk size, it may take time for all
chunks to split to the new size.
Splits cannot be undone. If you increase the chunk size, existing chunks must grow through inserts or updates
until they reach the new size.
Note: Chunk ranges are inclusive of the lower boundary and exclusive of the upper boundary.

Shard Key Indexes

All sharded collections must have an index that starts with the shard key. If you shard a collection without any
documents and without such an index, the shardCollection command will create the index on the shard key. If
the collection already has documents, you must create the index before using shardCollection.
Changed in version 2.2: The index on the shard key no longer needs to be only on the shard key. This index can be an
index of the shard key itself, or a compound index where the shard key is a prefix of the index.
Important: The index on the shard key cannot be a multikey index (page 324).
A sharded collection named people has for its shard key the field zipcode. It currently has the index {
zipcode: 1 }. You can replace this index with a compound index { zipcode: 1, username: 1 },
as follows:
1. Create an index on { zipcode:

1, username:

1 }:

db.people.ensureIndex( { zipcode: 1, username: 1 } );

2. When MongoDB finishes building the index, you can safely drop the existing index on { zipcode:

1 }:

db.people.dropIndex( { zipcode: 1 } );

Since the index on the shard key cannot be a multikey index, the index { zipcode: 1, username:
can only replace the index { zipcode: 1 } if there are no array values for the username field.

1 }

If you drop the last valid index for the shard key, recover by recreating an index on just the shard key.
For restrictions on shard key indexes, see limits-shard-keys.
Sharded Cluster Metadata
Config servers (page 488) store the metadata for a sharded cluster. The metadata reflects state and organization of the
sharded data sets and system. The metadata includes the list of chunks on every shard and the ranges that define the
chunks. The mongos instances cache this data and use it to route read and write operations to shards.
Config servers store the metadata in the Config Database (page 547).
Important: Always back up the config database before doing any maintenance on the config server.
To access the config database, issue the following command from the mongo shell:
use config

9.2. Sharding Concepts

505

MongoDB Documentation, Release 2.4.9

In general, you should never edit the content of the config database directly. The config database contains the
following collections:
changelog (page 548)
chunks (page 549)
collections (page 550)
databases (page 550)
lockpings (page 550)
locks (page 550)
mongos (page 551)
settings (page 551)
shards (page 552)
version (page 552)
For more information on these collections and their role in sharded clusters, see Config Database (page 547). See Read
and Write Operations on Config Servers (page 488) for more information about reads and updates to the metadata.

9.3 Sharded Cluster Tutorials

The following tutorials provide instructions for administering sharded clusters. For a higher-level overview, see Sharding (page 479).
Sharded Cluster Deployment Tutorials (page 507) Instructions for deploying sharded clusters, adding shards, selecting shard keys, and the initial configuration of sharded clusters.
Deploy a Sharded Cluster (page 507) Set up a sharded cluster by creating the needed data directories, starting
the required MongoDB instances, and configuring the cluster settings.
Considerations for Selecting Shard Keys (page 511) Choose the field that MongoDB uses to parse a collections documents for distribution over the clusters shards. Each shard holds documents with values within
a certain range.
Shard a Collection Using a Hashed Shard Key (page 513) Shard a collection based on hashes of a fields values in order to ensure even distribution over the collections shards.
Add Shards to a Cluster (page 514) Add a shard to add capacity to a sharded cluster.
Continue reading from Sharded Cluster Deployment Tutorials (page 507) for additional tutorials.
Sharded Cluster Maintenance Tutorials (page 522) Procedures and tasks for common operations on active sharded
clusters.
View Cluster Configuration (page 523) View status information about the clusters databases, shards, and
chunks.
Remove Shards from an Existing Sharded Cluster (page 534) Migrate a single shards data and remove the
shard.
Migrate Config Servers with Different Hostnames (page 524) Migrate a config server to a new system that
uses a new hostname. If possible, avoid changing the hostname and instead use the Migrate Config Servers
with the Same Hostname (page 524) procedure.
Manage Shard Tags (page 541) Use tags to associate specific ranges of shard key values with specific shards.
Continue reading from Sharded Cluster Maintenance Tutorials (page 522) for additional tutorials.
506

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Sharded Cluster Data Management (page 536) Practices that address common issues in managing large sharded
data sets.
Troubleshoot Sharded Clusters (page 545) Presents solutions to common issues and concerns relevant to the administration and use of sharded clusters. Refer to FAQ: MongoDB Diagnostics (page 587) for general diagnostic
information.

9.3.1 Sharded Cluster Deployment Tutorials

The following tutorials provide information on deploying sharded clusters.
Deploy a Sharded Cluster (page 507) Set up a sharded cluster by creating the needed data directories, starting the
required MongoDB instances, and configuring the cluster settings.
Considerations for Selecting Shard Keys (page 511) Choose the field that MongoDB uses to parse a collections documents for distribution over the clusters shards. Each shard holds documents with values within a certain range.
Shard a Collection Using a Hashed Shard Key (page 513) Shard a collection based on hashes of a fields values in
order to ensure even distribution over the collections shards.
Enable Authentication in a Sharded Cluster (page 513) Control access to a sharded cluster through a key file and
the keyFile setting on each of the clusters components.
Add Shards to a Cluster (page 514) Add a shard to add capacity to a sharded cluster.
Deploy Three Config Servers for Production Deployments (page 515) Convert a test deployment with one config
server to a production deployment with three config servers.
Convert a Replica Set to a Replicated Sharded Cluster (page 515) Convert a replica set to a sharded cluster in which
each shard is its own replica set.
Convert Sharded Cluster to Replica Set (page 521) Replace your sharded cluster with a single replica set.
Deploy a Sharded Cluster
Use the following sequence of tasks to deploy a sharded cluster:
Warning: Sharding and localhost Addresses
If you use either localhost or 127.0.0.1 as the hostname portion of any host identifier, for example as the
host argument to addShard or the value to the --configdb run time option, then you must use localhost
or 127.0.0.1 for all host settings for any MongoDB instances in the cluster. If you mix localhost addresses and
remote host address, MongoDB will error.

Start the Config Server Database Instances

The config server processes are mongod instances that store the clusters metadata. You designate a mongod as a
config server using the --configsvr option. Each config server stores a complete copy of the clusters metadata.
In production deployments, you must deploy exactly three config server instances, each running on different servers
to assure good uptime and data safety. In test environments, you can run all three instances on a single server.
Important: All members of a sharded cluster must be able to connect to all other members of a sharded cluster,
including all shards and all config servers. Ensure that the network and security systems including all interfaces and
firewalls, allow these connections.

9.3. Sharded Cluster Tutorials

507

MongoDB Documentation, Release 2.4.9

1. Create data directories for each of the three config server instances. By default, a config server stores its data
files in the /data/configdb directory. You can choose a different location. To create a data directory, issue a
command similar to the following:
mkdir /data/configdb

2. Start the three config server instances. Start each by issuing a command using the following syntax:
mongod --configsvr --dbpath <path> --port <port>

The default port for config servers is 27019. You can specify a different port. The following example starts a
config server using the default port and default data directory:
mongod --configsvr --dbpath /data/configdb --port 27019

For additional command options, see http://docs.mongodb.org/manualreference/program/mongod

or http://docs.mongodb.org/manualreference/configuration-options.
Note: All config servers must be running and available when you first initiate a sharded cluster.

Start the mongos Instances

The mongos instances are lightweight and do not require data directories. You can run a mongos instance on a
system that runs other cluster components, such as on an application server or a server running a mongod process. By
default, a mongos instance runs on port 27017.
When you start the mongos instance, specify the hostnames of the three config servers, either in the configuration file
or as command line parameters.
Tip
To avoid downtime, give each config server a logical DNS name (unrelated to the servers physical or virtual hostname). Without logical DNS names, moving or renaming a config server requires shutting down every mongod and
mongos instance in the sharded cluster.
To start a mongos instance, issue a command using the following syntax:
mongos --configdb <config server hostnames>

For example, to start a mongos that connects to config server instance running on the following hosts and on the
default ports:
cfg0.example.net
cfg1.example.net
cfg2.example.net
You would issue the following command:
mongos --configdb cfg0.example.net:27019,cfg1.example.net:27019,cfg2.example.net:27019

Each mongos in a sharded cluster must use the same configdb string, with identical host names listed in identical
order.
If you start a mongos instance with a string that does not exactly match the string used by the other mongos instances
in the cluster, the mongos return a Config Database String Error (page 545) error and refuse to start.

508

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Add Shards to the Cluster

A shard can be a standalone mongod or a replica set. In a production environment, each shard should be a replica set.
1. From a mongo shell, connect to the mongos instance. Issue a command using the following syntax:
mongo --host <hostname of machine running mongos> --port <port mongos listens on>

For example, if a mongos is accessible at mongos0.example.net on port 27017, issue the following
command:
mongo --host mongos0.example.net --port 27017

2. Add each shard to the cluster using the sh.addShard() method, as shown in the examples below. Issue
sh.addShard() separately for each shard. If the shard is a replica set, specify the name of the replica set and
specify a member of the set. In production deployments, all shards should be replica sets.
Optional
You can instead use the addShard database command, which lets you specify a name and maximum size for
the shard. If you do not specify these, MongoDB automatically assigns a name and maximum size. To use the
database command, see addShard.
The following are examples of adding a shard with sh.addShard():
To add a shard for a replica set named rs1 with a member running on port 27017 on
mongodb0.example.net, issue the following command:
sh.addShard( "rs1/mongodb0.example.net:27017" )

Changed in version 2.0.3.

For MongoDB versions prior to 2.0.3, you must specify all members of the replica set. For example:

sh.addShard( "rs1/mongodb0.example.net:27017,mongodb1.example.net:27017,mongodb2.example.net

To add a shard for a standalone mongod on port 27017 of mongodb0.example.net, issue the following command:
sh.addShard( "mongodb0.example.net:27017" )

Note: It might take some time for chunks to migrate to the new shard.

Enable Sharding for a Database

Before you can shard a collection, you must enable sharding for the collections database. Enabling sharding for a
database does not redistribute data but make it possible to shard the collections in that database.
Once you enable sharding for a database, MongoDB assigns a primary shard for that database where MongoDB stores
all data before sharding begins.
1. From a mongo shell, connect to the mongos instance. Issue a command using the following syntax:
mongo --host <hostname of machine running mongos> --port <port mongos listens on>

2. Issue the sh.enableSharding() method, specifying the name of the database for which to enable sharding.
Use the following syntax:

9.3. Sharded Cluster Tutorials

509

MongoDB Documentation, Release 2.4.9

sh.enableSharding("<database>")

Optionally, you can enable sharding for a database using the enableSharding command, which uses the following
syntax:
db.runCommand( { enableSharding: <database> } )

Enable Sharding for a Collection

You enable sharding on a per-collection basis.

1. Determine what you will use for the shard key. Your selection of the shard key affects the efficiency of sharding.
See the selection considerations listed in the Considerations for Selecting Shard Key (page 511).
2. If the collection already contains data you must create an index on the shard key using ensureIndex(). If
the collection is empty then MongoDB will create the index as part of the sh.shardCollection() step.
3. Enable sharding for a collection by issuing the sh.shardCollection() method in the mongo shell. The
method uses the following syntax:
sh.shardCollection("<database>.<collection>", shard-key-pattern)

Replace the <database>.<collection> string with the full namespace of your database, which consists
of the name of your database, a dot (e.g. .), and the full name of the collection. The shard-key-pattern
represents your shard key, which you specify in the same form as you would an index key pattern.
Example
The following sequence of commands shards four collections:
sh.shardCollection("records.people", { "zipcode": 1, "name": 1 } )
sh.shardCollection("people.addresses", { "state": 1, "_id": 1 } )
sh.shardCollection("assets.chairs", { "type": 1, "_id": 1 } )
db.alerts.ensureIndex( { _id : "hashed" } )
sh.shardCollection("events.alerts", { "_id": "hashed" } )

In order, these operations shard:

(a) The people collection in the records database using the shard key { "zipcode":
1 }.

1, "name":

This shard key distributes documents by the value of the zipcode field. If a number of documents have
the same value for this field, then that chunk will be splittable (page 512) by the values of the name field.
(b) The addresses collection in the people database using the shard key { "state":
1 }.

1, "_id":

This shard key distributes documents by the value of the state field. If a number of documents have the
same value for this field, then that chunk will be splittable (page 512) by the values of the _id field.
(c) The chairs collection in the assets database using the shard key { "type":
}.

1, "_id":

This shard key distributes documents by the value of the type field. If a number of documents have the
same value for this field, then that chunk will be splittable (page 512) by the values of the _id field.
(d) The alerts collection in the events database using the shard key { "_id":

"hashed" }.

New in version 2.4.

510

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

This shard key distributes documents by a hash of the value of the _id field. MongoDB computes the hash
of the _id field for the hashed index (page 343), which should provide an even distribution of documents
across a cluster.
Considerations for Selecting Shard Keys
Choosing a Shard Key

For many collections there may be no single, naturally occurring key that possesses all the qualities of a good shard
key. The following strategies may help construct a useful shard key from existing data:
1. Compute a more ideal shard key in your application layer, and store this in all of your documents, potentially in
the _id field.
2. Use a compound shard key that uses two or three values from all documents that provide the right mix of
cardinality with scalable write operations and query isolation.
3. Determine that the impact of using a less than ideal shard key is insignificant in your use case, given:
limited write volume,
expected data size, or
application query patterns.
4. New in version 2.4: Use a hashed shard key. Choose a field that has high cardinality and create a hashed index
(page 343) on that field. MongoDB uses these hashed index values as shard key values, which ensures an even
distribution of documents across the shards.
Tip
MongoDB automatically computes the hashes when resolving queries using hashed indexes. Applications do
not need to compute hashes.

Considerations for Selecting Shard Key Choosing the correct shard key can have a great impact on the performance, capability, and functioning of your database and cluster. Appropriate shard key choice depends on the schema
of your data and the way that your applications query and write data.
Create a Shard Key that is Easily Divisible

An easily divisible shard key makes it easy for MongoDB to distribute content among the shards. Shard keys that have
a limited number of possible values can result in chunks that are unsplittable.
See also:
Cardinality (page 512)
Create a Shard Key that has High Degree of Randomness

A shard key with high degree of randomness prevents any single shard from becoming a bottleneck and will distribute
write operations among the cluster.
See also:
Write Scaling (page 493)

9.3. Sharded Cluster Tutorials

511

MongoDB Documentation, Release 2.4.9

Create a Shard Key that Targets a Single Shard

A shard key that targets a single shard makes it possible for the mongos program to return most query operations
directly from a single specific mongod instance. Your shard key should be the primary field used by your queries.
Fields with a high degree of randomness make it difficult to target operations to specific shards.
See also:
Query Isolation (page 494)
Shard Using a Compound Shard Key

The challenge when selecting a shard key is that there is not always an obvious choice. Often, an existing field in your
collection may not be the optimal key. In those situations, computing a special purpose shard key into an additional
field or using a compound shard key may help produce one that is more ideal.
Cardinality

Cardinality in the context of MongoDB, refers to the ability of the system to partition data into chunks. For example,
consider a collection of data such as an address book that stores address records:
Consider the use of a state field as a shard key:
The state keys value holds the US state for a given address document. This field has a low cardinality as all
documents that have the same value in the state field must reside on the same shard, even if a particular states
chunk exceeds the maximum chunk size.
Since there are a limited number of possible values for the state field, MongoDB may distribute data unevenly
among a small number of fixed chunks. This may have a number of effects:
If MongoDB cannot split a chunk because all of its documents have the same shard key, migrations involving these un-splittable chunks will take longer than other migrations, and it will be more difficult for your
data to stay balanced.
If you have a fixed maximum number of chunks, you will never be able to use more than that number of
shards for this collection.
Consider the use of a zipcode field as a shard key:
While this field has a large number of possible values, and thus has potentially higher cardinality, its possible
that a large number of users could have the same value for the shard key, which would make this chunk of users
un-splittable.
In these cases, cardinality depends on the data. If your address book stores records for a geographically distributed contact list (e.g. Dry cleaning businesses in America,) then a value like zipcode would be sufficient.
However, if your address book is more geographically concentrated (e.g ice cream stores in Boston Massachusetts,) then you may have a much lower cardinality.
Consider the use of a phone-number field as a shard key:
Phone number has a high cardinality, because users will generally have a unique value for this field, MongoDB
will be able to split as many chunks as needed.
While high cardinality, is necessary for ensuring an even distribution of data, having a high cardinality does not
guarantee sufficient query isolation (page 494) or appropriate write scaling (page 493).

512

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Shard a Collection Using a Hashed Shard Key

New in version 2.4.
Hashed shard keys (page 493) use a hashed index (page 343) of a field as the shard key to partition data across your
sharded cluster.
For suggestions on choosing the right field as your hashed shard key, see Hashed Shard Keys (page 493). For limitations on hashed indexes, see Create a Hashed Index (page 343).
Note: If chunk migrations are in progress while creating a hashed shard key collection, the initial chunk distribution
may be uneven until the balancer automatically balances the collection.

Shard the Collection

To shard a collection using a hashed shard key, use an operation in the mongo that resembles the following:
sh.shardCollection( "records.active", { a: "hashed" } )

This operation shards the active collection in the records database, using a hash of the a field as the shard key.
Specify the Initial Number of Chunks

If you shard an empty collection using a hashed shard key, MongoDB automatically creates and migrates empty chunks
so that each shard has two chunks. To control how many chunks MongoDB creates when sharding the collection, use
shardCollection with the numInitialChunks parameter.
Important: MongoDB 2.4 adds support for hashed shard keys. After sharding a collection with a hashed shard key,
you must use the MongoDB 2.4 or higher mongos and mongod instances in your sharded cluster.
Warning: MongoDB hashed indexes truncate floating point numbers to 64-bit integers before hashing. For
example, a hashed index would store the same value for a field that held a value of 2.3, 2.2, and 2.9. To
prevent collisions, do not use a hashed index for floating point numbers that cannot be reliably converted to
64-bit integers (and then back to floating point). MongoDB hashed indexes do not support floating point values
larger than 253 .

Enable Authentication in a Sharded Cluster

New in version 2.0: Support for authentication with sharded clusters.
To control access to a sharded cluster, create key files and then set the keyFile option on all components of the
sharded cluster, including all mongos instances, all config server mongod instances, and all shard mongod instances.
The content of the key file is arbitrary but must be the same on all cluster members.
Note: For an overview of authentication, see Access Control (page 239). For an overview of security, see Security
(page 237).

Procedure

To enable authentication, do the following:

9.3. Sharded Cluster Tutorials

513

MongoDB Documentation, Release 2.4.9

1. Generate a key file to store authentication information, as described in the Generate a Key File (page 261)
section.
2. On each component in the sharded cluster, enable authentication by doing one of the following:
In the configuration file, set the keyFile option to the key files path and then start the component, as in
the following example:
keyFile = /srv/mongodb/keyfile

When starting the component, set --keyFile option, which is an option for both mongos instances and
mongod instances. Set the --keyFile to the key files path.
Note: The keyFile setting implies auth, which means in most cases you do not need to set auth explicitly.
3. Add the first administrative user and then add subsequent users. See Create a User Administrator (page 258).
Add Shards to a Cluster
You add shards to a sharded cluster after you create the cluster or anytime that you need to add capacity to the cluster.
If you have not created a sharded cluster, see Deploy a Sharded Cluster (page 507).
When adding a shard to a cluster, you should always ensure that the cluster has enough capacity to support the
migration without affecting legitimate production traffic.
In production environments, all shards should be replica sets.
Add a Shard to a Cluster

You interact with a sharded cluster by connecting to a mongos instance.

1. From a mongo shell, connect to the mongos instance. For example, if a mongos is accessible at
mongos0.example.net on port 27017, issue the following command:
mongo --host mongos0.example.net --port 27017

2. Add a shard to the cluster using the sh.addShard() method, as shown in the examples below. Issue
sh.addShard() separately for each shard. If the shard is a replica set, specify the name of the replica
set and specify a member of the set. In production deployments, all shards should be replica sets.
Optional
You can instead use the addShard database command, which lets you specify a name and maximum size for
the shard. If you do not specify these, MongoDB automatically assigns a name and maximum size. To use the
database command, see addShard.
The following are examples of adding a shard with sh.addShard():
To add a shard for a replica set named rs1 with a member running on port 27017 on
mongodb0.example.net, issue the following command:
sh.addShard( "rs1/mongodb0.example.net:27017" )

Changed in version 2.0.3.

For MongoDB versions prior to 2.0.3, you must specify all members of the replica set. For example:

514

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

sh.addShard( "rs1/mongodb0.example.net:27017,mongodb1.example.net:27017,mongodb2.example.net

To add a shard for a standalone mongod on port 27017 of mongodb0.example.net, issue the following command:
sh.addShard( "mongodb0.example.net:27017" )

Note: It might take some time for chunks to migrate to the new shard.

Deploy Three Config Servers for Production Deployments

This procedure converts a test deployment with only one config server (page 488) to a production deployment with
three config servers.
Tip
Use CNAMEs to identify your config servers to the cluster so that you can rename and renumber your config servers
without downtime.
For redundancy, all production sharded clusters (page 479) should deploy three config servers on three different
machines. Use a single config server only for testing deployments, never for production deployments. When you shift
to production, upgrade immediately to three config servers.
To convert a test deployment with one config server to a production deployment with three config servers:
1. Shut down all existing MongoDB processes in the cluster. This includes:
all mongod instances or replica sets that provide your shards.
all mongos instances in your cluster.
2. Copy the entire dbpath file system tree from the existing config server to the two machines that will provide the
additional config servers. These commands, issued on the system with the existing Config Database (page 547),
mongo-config0.example.net may resemble the following:
rsync -az /data/configdb mongo-config1.example.net:/data/configdb
rsync -az /data/configdb mongo-config2.example.net:/data/configdb

3. Start all three config servers, using the same invocation that you used for the single config server.
mongod --configsvr

4. Restart all shard mongod and mongos processes.

Convert a Replica Set to a Replicated Sharded Cluster
Overview

Following this tutorial, you will convert a single 3-member replica set to a cluster that consists of 2 shards. Each shard
will consist of an independent 3-member replica set.
The tutorial uses a test environment running on a local system UNIX-like system. You should feel encouraged to
follow along at home. If you need to perform this process in a production environment, notes throughout the
document indicate procedural differences.
The procedure, from a high level, is as follows:

9.3. Sharded Cluster Tutorials

515

MongoDB Documentation, Release 2.4.9

1. Create or select a 3-member replica set and insert some data into a collection.
2. Start the config databases and create a cluster with a single shard.
3. Create a second replica set with three new mongod instances.
4. Add the second replica set as a shard in the cluster.
5. Enable sharding on the desired collection or collections.
Process

Install MongoDB according to the instructions in the MongoDB Installation Tutorial (page 3).
Deploy a Replica Set with Test Data If have an existing MongoDB replica set deployment, you can omit the this
step and continue from Deploy Sharding Infrastructure (page 517).
Use the following sequence of steps to configure and deploy a replica set and to insert test data.
1. Create the following directories for the first replica set instance, named firstset:
/data/example/firstset1
/data/example/firstset2
/data/example/firstset3
To create directories, issue the following command:
mkdir -p /data/example/firstset1 /data/example/firstset2 /data/example/firstset3

2. In a separate terminal window or GNU Screen window, start three mongod instances by running each of the
following commands:
mongod --dbpath /data/example/firstset1 --port 10001 --replSet firstset --oplogSize 700 --rest
mongod --dbpath /data/example/firstset2 --port 10002 --replSet firstset --oplogSize 700 --rest
mongod --dbpath /data/example/firstset3 --port 10003 --replSet firstset --oplogSize 700 --rest

Note: The --oplogSize 700 option restricts the size of the operation log (i.e. oplog) for each mongod
instance to 700MB. Without the --oplogSize option, each mongod reserves approximately 5% of the free
disk space on the volume. By limiting the size of the oplog, each instance starts more quickly. Omit this setting
in production environments.
3. In a mongo shell session in a new terminal, connect to the mongodb instance on port 10001 by running the
following command. If you are in a production environment, first read the note below.
mongo localhost:10001/admin

Note: Above and hereafter, if you are running in a production environment or are testing this process with
mongod instances on multiple systems, replace localhost with a resolvable domain, hostname, or the IP
address of your system.
4. In the mongo shell, initialize the first replica set by issuing the following command:
db.runCommand({"replSetInitiate" :
{"_id" : "firstset", "members" : [{"_id" : 1, "host" : "localhost:10001"},
{"_id" : 2, "host" : "localhost:10002"},
{"_id" : 3, "host" : "localhost:10003"}
]}})

516

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

{
"info" : "Config now saved locally.
"ok" : 1
}

Should come online in about a minute.",

5. In the mongo shell, create and populate a new collection by issuing the following sequence of JavaScript
operations:

use test
switched to db test
people = ["Marc", "Bill", "George", "Eliot", "Matt", "Trey", "Tracy", "Greg", "Steve", "Kristina
for(var i=0; i<1000000; i++){
name = people[Math.floor(Math.random()*people.length)];
user_id = i;
boolean = [true, false][Math.floor(Math.random()*2)];
added_at = new Date();
number = Math.floor(Math.random()*10001);
db.test_collection.save({"name":name, "user_id":user_id, "boolean": boolean, "added_at":added_at
}

The above operations add one million documents to the collection test_collection. This can take several
minutes, depending on your system.
The script adds the documents in the following form:

{ "_id" : ObjectId("4ed5420b8fc1dd1df5886f70"), "name" : "Greg", "user_id" : 4, "boolean" : true, "ad

Deploy Sharding Infrastructure This procedure creates the three config databases that store the clusters metadata.
Note: For development and testing environments, a single config database is sufficient. In production environments,
use three config databases. Because config instances store only the metadata for the sharded cluster, they have minimal
resource requirements.
1. Create the following data directories for three config database instances:
/data/example/config1
/data/example/config2
/data/example/config3
Issue the following command at the system prompt:
mkdir -p /data/example/config1 /data/example/config2 /data/example/config3

2. In a separate terminal window or GNU Screen window, start the config databases by running the following
commands:
mongod --configsvr --dbpath /data/example/config1 --port 20001
mongod --configsvr --dbpath /data/example/config2 --port 20002
mongod --configsvr --dbpath /data/example/config3 --port 20003

3. In a separate terminal window or GNU Screen window, start mongos instance by running the following command:
mongos --configdb localhost:20001,localhost:20002,localhost:20003 --port 27017 --chunkSize 1

Note: If you are using the collection created earlier or are just experimenting with sharding, you can use a

9.3. Sharded Cluster Tutorials

517

MongoDB Documentation, Release 2.4.9

small --chunkSize (1MB works well.) The default chunkSize of 64MB means that your cluster must
have 64MB of data before the MongoDBs automatic sharding begins working.
In production environments, do not use a small shard size.
The configdb options specify the configuration databases (e.g.
localhost:20001,
localhost:20002, and localhost:2003). The mongos instance runs on the default MongoDB port (i.e. 27017), while the databases themselves are running on ports in the 30001 series. In the this
example, you may omit the --port 27017 option, as 27017 is the default port.
4. Add the first shard in mongos. In a new terminal window or GNU Screen session, add the first shard, according
to the following procedure:
(a) Connect to the mongos with the following command:
mongo localhost:27017/admin

(b) Add the first shard to the cluster by issuing the addShard command:
db.runCommand( { addShard : "firstset/localhost:10001,localhost:10002,localhost:10003" } )

(c) Observe the following message, which denotes success:

{ "shardAdded" : "firstset", "ok" : 1 }

Deploy a Second Replica Set This procedure deploys a second replica set. This closely mirrors the process used to
establish the first replica set above, omitting the test data.
1. Create the following data directories for the members of the second replica set, named secondset:
/data/example/secondset1
/data/example/secondset2
/data/example/secondset3
2. In three new terminal windows, start three instances of mongod with the following commands:

mongod --dbpath /data/example/secondset1 --port 10004 --replSet secondset --oplogSize 700 --rest
mongod --dbpath /data/example/secondset2 --port 10005 --replSet secondset --oplogSize 700 --rest
mongod --dbpath /data/example/secondset3 --port 10006 --replSet secondset --oplogSize 700 --rest

Note: As above, the second replica set uses the smaller oplogSize configuration. Omit this setting in
production environments.
3. In the mongo shell, connect to one mongodb instance by issuing the following command:
mongo localhost:10004/admin

4. In the mongo shell, initialize the second replica set by issuing the following command:
db.runCommand({"replSetInitiate" :
{"_id" : "secondset",
"members" : [{"_id" : 1, "host" : "localhost:10004"},
{"_id" : 2, "host" : "localhost:10005"},
{"_id" : 3, "host" : "localhost:10006"}
]}})
{
"info" : "Config now saved locally.

518

Should come online in about a minute.",

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

"ok" : 1
}

5. Add the second replica set to the cluster. Connect to the mongos instance created in the previous procedure and
issue the following sequence of commands:
use admin
db.runCommand( { addShard : "secondset/localhost:10004,localhost:10005,localhost:10006" } )

This command returns the following success message:

{ "shardAdded" : "secondset", "ok" : 1 }

6. Verify that both shards are properly configured by running the listShards command. View this and example
output below:
db.runCommand({listShards:1})
{
"shards" : [
{
"_id" : "firstset",
"host" : "firstset/localhost:10001,localhost:10003,localhost:10002"
},
{
"_id" : "secondset",
"host" : "secondset/localhost:10004,localhost:10006,localhost:10005"
}
],
"ok" : 1
}

Enable Sharding MongoDB must have sharding enabled on both the database and collection levels.
Enabling Sharding on the Database Level Issue the enableSharding command. The following example enables sharding on the test database:
db.runCommand( { enableSharding : "test" } )
{ "ok" : 1 }

Create an Index on the Shard Key MongoDB uses the shard key to distribute documents between shards. Once
selected, you cannot change the shard key. Good shard keys:
have values that are evenly distributed among all documents,
group documents that are often accessed at the same time into contiguous chunks, and
allow for effective distribution of activity among shards.
Typically shard keys are compound, comprising of some sort of hash and some sort of other primary key. Selecting
a shard key depends on your data set, application architecture, and usage pattern, and is beyond the scope of this
document. For the purposes of this example, we will shard the number key. This typically would not be a good
shard key for production deployments.
Create the index with the following procedure:
use test
db.test_collection.ensureIndex({number:1})

9.3. Sharded Cluster Tutorials

519

MongoDB Documentation, Release 2.4.9

See also:
The Shard Key Overview (page 492) and Shard Key (page 492) sections.
Shard the Collection Issue the following command:
use admin
db.runCommand( { shardCollection : "test.test_collection", key : {"number":1} })
{ "collectionsharded" : "test.test_collection", "ok" : 1 }

The collection test_collection is now sharded!

Over the next few minutes the Balancer begins to redistribute chunks of documents. You can confirm this activity by
switching to the test database and running db.stats() or db.printShardingStatus().
As clients insert additional documents into this collection, mongos distributes the documents evenly between the
shards.
In the mongo shell, issue the following commands to return statics against each cluster:
use test
db.stats()
db.printShardingStatus()

Example output of the db.stats() command:

{
"raw" : {
"firstset/localhost:10001,localhost:10003,localhost:10002" : {
"db" : "test",
"collections" : 3,
"objects" : 973887,
"avgObjSize" : 100.33173458522396,
"dataSize" : 97711772,
"storageSize" : 141258752,
"numExtents" : 15,
"indexes" : 2,
"indexSize" : 56978544,
"fileSize" : 1006632960,
"nsSizeMB" : 16,
"ok" : 1
},
"secondset/localhost:10004,localhost:10006,localhost:10005" : {
"db" : "test",
"collections" : 3,
"objects" : 26125,
"avgObjSize" : 100.33286124401914,
"dataSize" : 2621196,
"storageSize" : 11194368,
"numExtents" : 8,
"indexes" : 2,
"indexSize" : 2093056,
"fileSize" : 201326592,
"nsSizeMB" : 16,
"ok" : 1
}
},
"objects" : 1000012,
"avgObjSize" : 100.33176401883178,
"dataSize" : 100332968,

520

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

"storageSize" : 152453120,
"numExtents" : 23,
"indexes" : 4,
"indexSize" : 59071600,
"fileSize" : 1207959552,
"ok" : 1
}

Example output of the db.printShardingStatus() command:

--- Sharding Status --sharding version: { "_id" : 1, "version" : 3 }
shards:
{ "_id" : "firstset", "host" : "firstset/localhost:10001,localhost:10003,localhost:10002" }
{ "_id" : "secondset", "host" : "secondset/localhost:10004,localhost:10006,localhost:10005" }
databases:
{ "_id" : "admin", "partitioned" : false, "primary" : "config" }
{ "_id" : "test", "partitioned" : true, "primary" : "firstset" }
test.test_collection chunks:
secondset
5
firstset
186
[...]

In a few moments you can run these commands for a second time to demonstrate that chunks are migrating from
firstset to secondset.
When this procedure is complete, you will have converted a replica set into a cluster where each shard is itself a replica
set.
Convert Sharded Cluster to Replica Set
This tutorial describes the process for converting a sharded cluster to a non-sharded replica set. To convert a replica set
into a sharded cluster Convert a Replica Set to a Replicated Sharded Cluster (page 515). See the Sharding (page 479)
documentation for more information on sharded clusters.
Convert a Cluster with a Single Shard into a Replica Set

In the case of a sharded cluster with only one shard, that shard contains the full data set. Use the following procedure
to convert that cluster into a non-sharded replica set:
1. Reconfigure the application to connect to the primary member of the replica set hosting the single shard that
system will be the new replica set.
2. Optionally remove the --shardsrv option, if your mongod started with this option.
Tip
Changing the --shardsrv option will change the port that mongod listens for incoming connections on.
The single-shard cluster is now a non-sharded replica set that will accept read and write operations on the data set.
You may now decommission the remaining sharding infrastructure.

9.3. Sharded Cluster Tutorials

521

MongoDB Documentation, Release 2.4.9

Convert a Sharded Cluster into a Replica Set

Use the following procedure to transition from a sharded cluster with more than one shard to an entirely new replica
set.
1. With the sharded cluster running, deploy a new replica set (page 420) in addition to your sharded cluster. The
replica set must have sufficient capacity to hold all of the data files from all of the current shards combined. Do
not configure the application to connect to the new replica set until the data transfer is complete.
2. Stop all writes to the sharded cluster. You may reconfigure your application or stop all mongos instances.
If you stop all mongos instances, the applications will not be able to read from the database. If you stop all
mongos instances, start a temporary mongos instance on that applications cannot access for the data migration
procedure.
3. Use mongodump and mongorestore (page 195) to migrate the data from the mongos instance to the new replica
set.
Note: Not all collections on all databases are necessarily sharded. Do not solely migrate the sharded collections.
Ensure that all databases and all collections migrate correctly.
4. Reconfigure the application to use the non-sharded replica set instead of the mongos instance.
The application will now use the un-sharded replica set for reads and writes. You may now decommission the remaining unused sharded cluster infrastructure.

9.3.2 Sharded Cluster Maintenance Tutorials

The following tutorials provide information in maintaining sharded clusters.
View Cluster Configuration (page 523) View status information about the clusters databases, shards, and chunks.
Migrate Config Servers with the Same Hostname (page 524) Migrate a config server to a new system while keeping
the same hostname. This procedure requires changing the DNS entry to point to the new system.
Migrate Config Servers with Different Hostnames (page 524) Migrate a config server to a new system that uses a
new hostname. If possible, avoid changing the hostname and instead use the Migrate Config Servers with the
Same Hostname (page 524) procedure.
Replace a Config Server (page 525) Replaces a config server that has become inoperable. This procedure assumes
that the hostname does not change.
Migrate a Sharded Cluster to Different Hardware (page 526) Migrate a sharded cluster to a different hardware system, for example, when moving a pre-production environment to production.
Backup Cluster Metadata (page 529) Create a backup of a sharded clusters metadata while keeping the cluster operational.
Configure Behavior of Balancer Process in Sharded Clusters (page 529) Manage the balancers behavior by
scheduling a balancing window, changing size settings, or requiring replication before migration.
Manage Sharded Cluster Balancer (page 531) View balancer status and manage balancer behavior.
Remove Shards from an Existing Sharded Cluster (page 534) Migrate a single shards data and remove the shard.

522

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

View Cluster Configuration

List Databases with Sharding Enabled

To list the databases that have sharding enabled, query the databases collection in the Config Database (page 547).
A database has sharding enabled if the value of the partitioned field is true. Connect to a mongos instance
with a mongo shell, and run the following operation to get a full list of databases with sharding enabled:
use config
db.databases.find( { "partitioned": true } )

Example
You can use the following sequence of commands when to return a list of all databases in the cluster:
use config
db.databases.find()

If this returns the following result set:

{ "_id" : "admin", "partitioned" : false, "primary" : "config" }
{ "_id" : "animals", "partitioned" : true, "primary" : "m0.example.net:30001" }
{ "_id" : "farms", "partitioned" : false, "primary" : "m1.example2.net:27017" }

Then sharding is only enabled for the animals database.

List Shards

To list the current set of configured shards, use the listShards command, as follows:
use admin
db.runCommand( { listShards : 1 } )

View Cluster Details

To view cluster details, issue db.printShardingStatus() or sh.status(). Both methods return the same
output.
Example
In the following example output from sh.status()
sharding version displays the version number of the shard metadata.
shards displays a list of the mongod instances used as shards in the cluster.
databases displays all databases in the cluster, including database that do not have sharding enabled.
The chunks information for the foo database displays how many chunks are on each shard and displays the
range of each chunk.
--- Sharding Status --sharding version: { "_id" : 1, "version" : 3 }
shards:
{ "_id" : "shard0000", "host" : "m0.example.net:30001" }
{ "_id" : "shard0001", "host" : "m3.example2.net:50000" }
databases:

9.3. Sharded Cluster Tutorials

523

MongoDB Documentation, Release 2.4.9

{ "_id" : "admin", "partitioned" : false, "primary" : "config" }

{ "_id" : "contacts", "partitioned" : true, "primary" : "shard0000" }
foo.contacts
shard key: { "zip" : 1 }
chunks:
shard0001
2
shard0002
3
shard0000
2
{ "zip" : { "$minKey" : 1 } } -->> { "zip" : 56000 } on : shard0001 { "t" : 2, "i" : 0 }
{ "zip" : 56000 } -->> { "zip" : 56800 } on : shard0002 { "t" : 3, "i" : 4 }
{ "zip" : 56800 } -->> { "zip" : 57088 } on : shard0002 { "t" : 4, "i" : 2 }
{ "zip" : 57088 } -->> { "zip" : 57500 } on : shard0002 { "t" : 4, "i" : 3 }
{ "zip" : 57500 } -->> { "zip" : 58140 } on : shard0001 { "t" : 4, "i" : 0 }
{ "zip" : 58140 } -->> { "zip" : 59000 } on : shard0000 { "t" : 4, "i" : 1 }
{ "zip" : 59000 } -->> { "zip" : { "$maxKey" : 1 } } on : shard0000 { "t" : 3, "i" : 3 }
{ "_id" : "test", "partitioned" : false, "primary" : "shard0000" }

Migrate Config Servers with the Same Hostname

This procedure migrates a config server (page 488) in a sharded cluster (page 484) to a new system that uses the same
hostname.
To migrate all the config servers in a cluster, perform this procedure for each config server separately and migrate the
config servers in reverse order from how they are listed in the mongos instances configdb string. Start with the
last config server listed in the configdb string.
1. Shut down the config server.
This renders all config data for the sharded cluster read only.
2. Change the DNS entry that points to the system that provided the old config server, so that the same hostname
points to the new system. How you do this depends on how you organize your DNS and hostname resolution
services.
3. Copy the contents of dbpath from the old config server to the new config server.
For example, to copy the contents of dbpath to a machine named mongodb.config2.example.net,
you might issue a command similar to the following:
rsync -az /data/configdb/ mongodb.config2.example.net:/data/configdb

4. Start the config server instance on the new system. The default invocation is:
mongod --configsvr

When you start the third config server, your cluster will become writable and it will be able to create new splits and
migrate chunks as needed.
Migrate Config Servers with Different Hostnames
This procedure migrates a config server (page 488) in a sharded cluster (page 484) to a new server that uses a different
hostname. Use this procedure only if the config server will not be accessible via the same hostname.
Changing a config servers (page 488) hostname requires downtime and requires restarting every process in the
sharded cluster. If possible, avoid changing the hostname so that you can instead use the procedure to migrate a config
server and use the same hostname (page 524).

524

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

To migrate all the config servers in a cluster, perform this procedure for each config server separately and migrate the
config servers in reverse order from how they are listed in the mongos instances configdb string. Start with the
last config server listed in the configdb string.
1. Disable the cluster balancer process temporarily. See Disable the Balancer (page 533) for more information.
2. Shut down the config server.
This renders all config data for the sharded cluster read only.
3. Copy the contents of dbpath from the old config server to the new config server.
Example
To copy the contents of dbpath to a machine named mongodb.config2.example.net, use a command
that resembles the following:
rsync -az /data/configdb mongodb.config2.example.net:/data/configdb

4. Start the config server instance on the new system. The default invocation is:
mongod --configsvr

5. Shut down all existing MongoDB processes. This includes:

the mongod instances or replica sets that provide your shards.
the mongod instances that provide your existing config databases (page 547).
the mongos instances.
6. Restart all mongod processes that provide the shard servers.
7. Update the configdb setting for each mongos instances.
8. Restart the mongos instances.
9. Re-enable the balancer to allow the cluster to resume normal balancing operations. See the Disable the Balancer
(page 533) section for more information on managing the balancer process.
Replace a Config Server
This procedure replaces an inoperable config server (page 488) in a sharded cluster (page 484). Use this procedure
only to replace a config server that has become inoperable (e.g. hardware failure).
This process assumes that the hostname of the instance will not change. If you must change the hostname of the
instance, use the procedure to migrate a config server and use a new hostname (page 524).
1. Disable the cluster balancer process temporarily. See Disable the Balancer (page 533) for more information.
2. Provision a new system, with the same hostname as the previous host.
You will have to ensure that the new system has the same IP address and hostname as the system its replacing
or you will need to modify the DNS records and wait for them to propagate.
3. Shut down one (and only one) of the existing config servers. Copy all of this hosts dbpath file system tree
from the current system to the system that will provide the new config server. This command, issued on the
system with the data files, may resemble the following:
rsync -az /data/configdb mongodb.config2.example.net:/data/configdb

4. Restart the config server process that you used in the previous step to copy the data files to the new config server
instance.
9.3. Sharded Cluster Tutorials

525

MongoDB Documentation, Release 2.4.9

5. Start the new config server instance. The default invocation is:
mongod --configsvr

6. Re-enable the balancer to allow the cluster to resume normal balancing operations. See the Disable the Balancer
(page 533) section for more information on managing the balancer process.
Note: In the course of this procedure never remove a config server from the configdb parameter on any of the
mongos instances. If you need to change the name of a config server, always make sure that all mongos instances
have three config servers specified in the configdb setting at all times.

Migrate a Sharded Cluster to Different Hardware

This procedure moves the components of the sharded cluster to a new hardware system without downtime for reads
and writes.
Important: While the migration is in progress, do not attempt to change to the cluster metadata (page 505). Do not
use any operation that modifies the cluster metadata in any way. For example, do not create or drop databases, create
or drop collections, or use any sharding commands.
If your cluster includes a shard backed by a standalone mongod instance, consider converting the standalone to a
replica set (page 432) to simplify migration and to let you keep the cluster online during future maintenance. Migrating
a shard as standalone is a multi-step process that may require downtime.
To migrate a cluster to new hardware, perform the following tasks.
Disable the Balancer

Disable the balancer to stop chunk migration (page 502) and do not perform any metadata write operations until the
process finishes. If a migration is in progress, the balancer will complete the in-progress migration before stopping.
To disable the balancer, connect to one of the clusters mongos instances and issue the following method:
sh.stopBalancer()

To check the balancer state, issue the sh.getBalancerState() method.

For more information, see Disable the Balancer (page 533).
Migrate Each Config Server Separately

Migrate each config server (page 488) by starting with the last config server listed in the configdb string. Proceed
in reverse order of the configdb string. Migrate and restart a config server before proceeding to the next. Do not
rename a config server during this process.
Note: If the name or address that a sharded cluster uses to connect to a config server changes, you must restart every
mongod and mongos instance in the sharded cluster. Avoid downtime by using CNAMEs to identify config servers
within the MongoDB deployment.
See Migrate Config Servers with Different Hostnames (page 524) for more information.
Important: Start with the last config server listed in configdb.

526

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

1. Shut down the config server.

This renders all config data for the sharded cluster read only.
2. Change the DNS entry that points to the system that provided the old config server, so that the same hostname
points to the new system. How you do this depends on how you organize your DNS and hostname resolution
services.
3. Copy the contents of dbpath from the old config server to the new config server.
For example, to copy the contents of dbpath to a machine named mongodb.config2.example.net,
you might issue a command similar to the following:
rsync -az /data/configdb/ mongodb.config2.example.net:/data/configdb

4. Start the config server instance on the new system. The default invocation is:
mongod --configsvr

Restart the mongos Instances

If the configdb string will change as part of the migration, you must shut down all mongos instances before
changing the configdb string. This avoids errors in the sharded cluster over configdb string conflicts.
If the configdb string will remain the same, you can migrate the mongos instances sequentially or all at once.
1. Shut down the mongos instances using the shutdown command. If the configdb string is changing, shut
down all mongos instances.
2. If the hostname has changed for any of the config servers, update the configdb string for each mongos
instance. The mongos instances must all use the same configdb string. The strings must list identical host
names in identical order.
Tip
To avoid downtime, give each config server a logical DNS name (unrelated to the servers physical or virtual
hostname). Without logical DNS names, moving or renaming a config server requires shutting down every
mongod and mongos instance in the sharded cluster.
3. Restart the mongos instances being sure to use the updated configdb string if hostnames have changed.
For more information, see Start the mongos Instances (page 508).
Migrate the Shards

Migrate the shards one at a time. For each shard, follow the appropriate procedure in this section.
Migrate a Replica Set Shard To migrate a sharded cluster, migrate each member separately. First migrate the
non-primary members, and then migrate the primary last.
If the replica set has two voting members, add an arbiter (page 389) to the replica set to ensure the set keeps a majority
of its votes available during the migration. You can remove the arbiter after completing the migration.

9.3. Sharded Cluster Tutorials

527

MongoDB Documentation, Release 2.4.9

Migrate a Member of a Replica Set Shard

1. Shut down the mongod process. To ensure a clean shutdown, use the shutdown command.
2. Move the data directory (i.e., the dbpath) to the new machine.
3. Restart the mongod process at the new location.
4. Connect to the replica sets current primary.
5. If the hostname of the member has changed, use rs.reconfig() to update the replica set configuration
document (page 467) with the new hostname.
For example, the following sequence of commands updates the hostname for the instance at position 2 in the
members array:
cfg = rs.conf()
cfg.members[2].host = "pocatello.example.net:27017"
rs.reconfig(cfg)

For more information on updating the configuration document, see Example Reconfiguration Operations
(page 470).
6. To confirm the new configuration, issue rs.conf().
7. Wait for the member to recover. To check the members state, issue rs.status().
Migrate the Primary in a Replica Set Shard While migrating the replica sets primary, the set must elect a new
primary. This failover process which renders the replica set unavailable to perform reads or accept writes for the
duration of the election, which typically completes quickly. If possible, plan the migration during a maintenance
window.
1. Step down the primary to allow the normal failover (page 397) process. To step down the primary, connect to
the primary and issue the either the replSetStepDown command or the rs.stepDown() method. The
following example shows the rs.stepDown() method:
rs.stepDown()

2. Once the primary has stepped down and another member has become PRIMARY (page 474) state. To migrate
the stepped-down primary, follow the Migrate a Member of a Replica Set Shard (page 528) procedure
You can check the output of rs.status() to confirm the change in status.
Migrate a Standalone Shard The ideal procedure for migrating a standalone shard is to convert the standalone to a
replica set (page 432) and then use the procedure for migrating a replica set shard (page 527). In production clusters,
all shards should be replica sets, which provides continued availability during maintenance windows.
Migrating a shard as standalone is a multi-step process during which part of the shard may be unavailable. If the shard
is the primary shard for a database,the process includes the movePrimary command. While the movePrimary
runs, you should stop modifying data in that database. To migrate the standalone shard, use the Remove Shards from
an Existing Sharded Cluster (page 534) procedure.
Re-Enable the Balancer

To complete the migration, re-enable the balancer to resume chunk migrations (page 502).
Connect to one of the clusters mongos instances and pass true to the sh.setBalancerState() method:

528

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

sh.setBalancerState(true)

To check the balancer state, issue the sh.getBalancerState() method.

For more information, see Enable the Balancer (page 533).
Backup Cluster Metadata
This procedure shuts down the mongod instance of a config server (page 488) in order to create a backup of a sharded
clusters (page 479) metadata. The clusters config servers store all of the clusters metadata, most importantly the
mapping from chunks to shards.
When you perform this procedure, the cluster remains operational 8 .
1. Disable the cluster balancer process temporarily. See Disable the Balancer (page 533) for more information.
2. Shut down one of the config databases.
3. Create a full copy of the data files (i.e. the path specified by the dbpath option for the config instance.)
4. Restart the original configuration server.
5. Re-enable the balancer to allow the cluster to resume normal balancing operations. See the Disable the Balancer
(page 533) section for more information on managing the balancer process.
See also:
MongoDB Backup Methods (page 136).
Configure Behavior of Balancer Process in Sharded Clusters
The balancer is a process that runs on one of the mongos instances in a cluster and ensures that chunks are evenly
distributed throughout a sharded cluster. In most deployments, the default balancer configuration is sufficient for
normal operation. However, administrators might need to modify balancer behavior depending on application or
operational requirements. If you encounter a situation where you need to modify the behavior of the balancer, use the
procedures described in this document.
For conceptual information about the balancer, see Sharded Collection Balancing (page 501) and Cluster Balancer
(page 501).
Schedule a Window of Time for Balancing to Occur

You can schedule a window of time during which the balancer can migrate chunks, as described in the following
procedures:
Schedule the Balancing Window (page 532)
Remove a Balancing Window Schedule (page 532).
The mongos instances user their own local timezones to when respecting balancer window.
8 While one of the three config servers is unavailable, the cluster cannot split any chunks nor can it migrate chunks between shards. Your
application will be able to write data to the cluster. See Config Servers (page 488) for more information.

9.3. Sharded Cluster Tutorials

529

MongoDB Documentation, Release 2.4.9

Configure Default Chunk Size

The default chunk size for a sharded cluster is 64 megabytes. In most situations, the default size is appropriate for
splitting and migrating chunks. For information on how chunk size affects deployments, see details, see Chunk Size
(page 504).
Changing the default chunk size affects chunks that are processes during migrations and auto-splits but does not
retroactively affect all chunks.
To configure default chunk size, see Modify Chunk Size in a Sharded Cluster (page 539).
Change the Maximum Storage Size for a Given Shard

The maxSize field in the shards (page 552) collection in the config database (page 547) sets the maximum size
for a shard, allowing you to control whether the balancer will migrate chunks to a shard. If mapped size 9 is above
a shards maxSize, the balancer will not move chunks to the shard. Also, the balancer will not move chunks off an
overloaded shard. This must happen manually. The maxSize value only affects the balancers selection of destination
shards.
By default, maxSize is not specified, allowing shards to consume the total amount of available space on their machines if necessary.
You can set maxSize both when adding a shard and once a shard is running.
To set maxSize when adding a shard, set the addShard commands maxSize parameter to the maximum size in
megabytes. For example, the following command run in the mongo shell adds a shard with a maximum size of 125
megabytes:
db.runCommand( { addshard : "example.net:34008", maxSize : 125 } )

To set maxSize on an existing shard, insert or update the maxSize field in the shards (page 552) collection in the
config database (page 547). Set the maxSize in megabytes.
Example
Assume you have the following shard without a maxSize field:
{ "_id" : "shard0000", "host" : "example.net:34001" }

Run the following sequence of commands in the mongo shell to insert a maxSize of 125 megabytes:
use config
db.shards.update( { _id : "shard0000" }, { $set : { maxSize : 125 } } )

To later increase the maxSize setting to 250 megabytes, run the following:
use config
db.shards.update( { _id : "shard0000" }, { $set : { maxSize : 250 } } )

Require Replication before Chunk Migration (Secondary Throttle)

New in version 2.2.1: _secondaryThrottle became an option to the balancer and to command moveChunk.
_secondaryThrottle makes it possible to require the balancer wait for replication to secondaries during migrations.
9

This value includes the mapped size of all data files including thelocal and admin databases. Account for this when setting maxSize.

530

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Changed in version 2.4: _secondaryThrottle became the default mode for all balancer and moveChunk operations.
Before 2.2.1, the write operations required to migrate chunks between shards do not need to replicate to secondaries in
order to succeed. However, you can configure the balancer to require migration related write operations to replicate to
secondaries. This throttles or slows the migration process and in doing so reduces the potential impact of migrations
on a sharded cluster.
You can throttle migrations by enabling the balancers _secondaryThrottle parameter. When enabled, secondary throttle requires a { w : 2 } write concern on delete and insertion operations, so that every operation
propagates to at least one secondary before the balancer issues the next operation.
Starting with version 2.4 the default secondaryThrottle value is true.
_secondaryThrottle to false.

To revert to previous behavior, set

You enable or disable _secondaryThrottle directly in the settings (page 551) collection in the config
database (page 547) by running the following commands from a mongo shell, connected to a mongos instance:
use config
db.settings.update( { "_id" : "balancer" } , { $set : { "_secondaryThrottle" : true } } , { upsert :

You also can enable secondary throttle when issuing the moveChunk command by setting _secondaryThrottle
to true. For more information, see moveChunk.
Manage Sharded Cluster Balancer
This page describes common administrative procedures related to balancing. For an introduction to balancing, see
Sharded Collection Balancing (page 501). For lower level information on balancing, see Cluster Balancer (page 501).
See also:
Configure Behavior of Balancer Process in Sharded Clusters (page 529)
Check the Balancer State

The following command checks if the balancer is enabled (i.e. that the balancer is allowed to run). The command does
not check if the balancer is active (i.e. if it is actively balancing chunks).
To see if the balancer is enabled in your cluster, issue the following command, which returns a boolean:
sh.getBalancerState()

Check the Balancer Lock

To see if the balancer process is active in your cluster, do the following:

1. Connect to any mongos in the cluster using the mongo shell.
2. Issue the following command to switch to the Config Database (page 547):
use config

3. Use the following query to return the balancer lock:

db.locks.find( { _id : "balancer" } ).pretty()

When this command returns, you will see output like the following:

9.3. Sharded Cluster Tutorials

531

MongoDB Documentation, Release 2.4.9

{
"_id" : "balancer",
"process" : "mongos0.example.net:1292810611:1804289383",
"state" : 2,
"ts" : ObjectId("4d0f872630c42d1978be8a2e"),
"when" : "Mon Dec 20 2010 11:41:10 GMT-0500 (EST)",
"who" : "mongos0.example.net:1292810611:1804289383:Balancer:846930886",
"why" : "doing balance round" }

This output confirms that:

The balancer originates from
mongos0.example.net.

the

mongos

running

on

the

system

with

the

hostname

The value in the state field indicates that a mongos has the lock. For version 2.0 and later, the value of an
active lock is 2; for earlier versions the value is 1.
Schedule the Balancing Window

In some situations, particularly when your data set grows slowly and a migration can impact performance, its useful
to be able to ensure that the balancer is active only at certain times. Use the following procedure to specify a window
during which the balancer will be able to migrate chunks:
1. Connect to any mongos in the cluster using the mongo shell.
2. Issue the following command to switch to the Config Database (page 547):
use config

3. Use an operation modeled on the following example update() operation to modify the balancers window:

db.settings.update({ _id : "balancer" }, { $set : { activeWindow : { start : "<start-time>", sto

Replace <start-time> and <end-time> with time values using two digit hour and minute values (e.g
HH:MM) that describe the beginning and end boundaries of the balancing window. These times will be evaluated
relative to the time zone of each individual mongos instance in the sharded cluster. If your mongos instances
are physically located in different time zones, use a common time zone (e.g. GMT) to ensure that the balancer
window is interpreted correctly.
For instance, running the following will force the balancer to run between 11PM and 6AM local time only:

db.settings.update({ _id : "balancer" }, { $set : { activeWindow : { start : "23:00", stop : "6:

Note: The balancer window must be sufficient to complete the migration of all data inserted during the day.
As data insert rates can change based on activity and usage patterns, it is important to ensure that the balancing window
you select will be sufficient to support the needs of your deployment.

Remove a Balancing Window Schedule

If you have set the balancing window (page 532) and wish to remove the schedule so that the balancer is always
running, issue the following sequence of operations:
use config
db.settings.update({ _id : "balancer" }, { $unset : { activeWindow : true } })

532

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Disable the Balancer

By default the balancer may run at any time and only moves chunks as needed. To disable the balancer for a short
period of time and prevent all migration, use the following procedure:
1. Connect to any mongos in the cluster using the mongo shell.
2. Issue the following operation to disable the balancer:
sh.setBalancerState(false)

If a migration is in progress, the system will complete the in-progress migration before stopping.
3. To verify that the balancer has stopped, issue the following command, which returns false if the balancer is
stopped:
sh.getBalancerState()

Optionally, to verify no migrations are in progress after disabling, issue the following operation in the mongo
shell:
use config
while( sh.isBalancerRunning() ) {
print("waiting...");
sleep(1000);
}

Note: To disable the balancer from a driver that does not have the sh.startBalancer() helper, issue the
following command from the config database:
db.settings.update( { _id: "balancer" }, { $set : { stopped: true } } , true )

Enable the Balancer

Use this procedure if you have disabled the balancer and are ready to re-enable it:
1. Connect to any mongos in the cluster using the mongo shell.
2. Issue one of the following operations to enable the balancer:
From the mongo shell, issue:
sh.setBalancerState(true)

From a driver that does not have the sh.startBalancer() helper, issue the following from the config
database:
db.settings.update( { _id: "balancer" }, { $set : { stopped: false } } , true )

Disable Balancing During Backups

If MongoDB migrates a chunk during a backup (page 136), you can end with an inconsistent snapshot of your sharded
cluster. Never run a backup while the balancer is active. To ensure that the balancer is inactive during your backup
operation:
Set the balancing window (page 532) so that the balancer is inactive during the backup. Ensure that the backup
can complete while you have the balancer disabled.

9.3. Sharded Cluster Tutorials

533

MongoDB Documentation, Release 2.4.9

manually disable the balancer (page 533) for the duration of the backup procedure.
If you turn the balancer off while it is in the middle of a balancing round, the shut down is not instantaneous. The
balancer completes the chunk move in-progress and then ceases all further balancing rounds.
Before starting a backup operation, confirm that the balancer is not active. You can use the following command to
determine if the balancer is active:
!sh.getBalancerState() && !sh.isBalancerRunning()

When the backup procedure is complete you can reactivate the balancer process.
Remove Shards from an Existing Sharded Cluster
To remove a shard you must ensure the shards data is migrated to the remaining shards in the cluster. This procedure
describes how to safely migrate data and how to remove a shard.
This procedure describes how to safely remove a single shard. Do not use this procedure to migrate an entire cluster
to new hardware. To migrate an entire shard to new hardware, migrate individual shards as if they were independent
replica sets.
To remove a shard, first connect to one of the clusters mongos instances using mongo shell. Then use the sequence
of tasks in this document to remove a shard from the cluster.
Ensure the Balancer Process is Enabled

To successfully migrate data from a shard, the balancer process must be enabled. Check the balancer state using
the sh.getBalancerState() helper in the mongo shell. For more information, see the section on balancer
operations (page 533).
Determine the Name of the Shard to Remove

To determine the name of the shard, connect to a mongos instance with the mongo shell and either:
Use the listShards command, as in the following:
db.adminCommand( { listShards: 1 } )

Run either the sh.status() or the db.printShardingStatus() method.

The shards._id field lists the name of each shard.
Remove Chunks from the Shard

Run the removeShard command. This begins draining chunks from the shard you are removing to other shards
in the cluster. For example, for a shard named mongodb0, run:
db.runCommand( { removeShard: "mongodb0" } )

This operation returns immediately, with the following response:

{ msg : "draining started successfully" , state: "started" , shard :"mongodb0" , ok : 1 }

Depending on your network capacity and the amount of data, this operation can take from a few minutes to several
days to complete.

534

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Check the Status of the Migration

To check the progress of the migration at any stage in the process, run removeShard. For example, for a shard
named mongodb0, run:
db.runCommand( { removeShard: "mongodb0" } )

The command returns output similar to the following:

{ msg: "draining ongoing" , state: "ongoing" , remaining: { chunks: NumberLong(42), dbs : NumberLong(

In the output, the remaining document displays the remaining number of chunks that MongoDB must migrate to
other shards and the number of MongoDB databases that have primary status on this shard.
Continue checking the status of the removeShard command until the number of chunks remaining is 0. Then proceed
to the next step.
Move Unsharded Data

If the shard is the primary shard for one or more databases in the cluster, then the shard will have unsharded data. If
the shard is not the primary shard for any databases, skip to the next task, Finalize the Migration (page 535).
In a cluster, a database with unsharded collections stores those collections only on a single shard. That shard becomes
the primary shard for that database. (Different databases in a cluster can have different primary shards.)
Warning: Do not perform this procedure until you have finished draining the shard.
1. To determine if the shard you are removing is the primary shard for any of the clusters databases, issue one of
the following methods:
sh.status()
db.printShardingStatus()
In the resulting document, the databases field lists each database and its primary shard. For example, the
following database field shows that the products database uses mongodb0 as the primary shard:
{

"_id" : "products",

"partitioned" : true,

"primary" : "mongodb0" }

2. To move a database to another shard, use the movePrimary command. For example, to migrate all remaining
unsharded data from mongodb0 to mongodb1, issue the following command:
db.runCommand( { movePrimary: "products", to: "mongodb1" })

This command does not return until MongoDB completes moving all data, which may take a long time. The
response from this command will resemble the following:
{ "primary" : "mongodb1", "ok" : 1 }

Finalize the Migration

To clean up all metadata information and finalize the removal, run removeShard again. For example, for a shard
named mongodb0, run:
db.runCommand( { removeShard: "mongodb0" } )

A success message appears at completion:

9.3. Sharded Cluster Tutorials

535

MongoDB Documentation, Release 2.4.9

{ msg: "remove shard completed successfully" , state: "completed", host: "mongodb0", ok : 1 }

Once the value of the stage field is completed, you may safely stop the processes comprising the mongodb0
shard.
See also:
Backup and Restore Sharded Clusters (page 198)

9.3.3 Sharded Cluster Data Management

The following documents provide information in managing data in sharded clusters.
Create Chunks in a Sharded Cluster (page 536) Create chunks, or pre-split empty collection to ensure an even distribution of chunks during data ingestion.
Split Chunks in a Sharded Cluster (page 537) Manually create chunks in a sharded collection.
Migrate Chunks in a Sharded Cluster (page 538) Manually migrate chunks without using the automatic balance
process.
Modify Chunk Size in a Sharded Cluster (page 539) Modify the default chunk size in a sharded collection
Tag Aware Sharding (page 540) Tags associate specific ranges of shard key values with specific shards for use in
managing deployment patterns.
Manage Shard Tags (page 541) Use tags to associate specific ranges of shard key values with specific shards.
Enforce Unique Keys for Sharded Collections (page 542) Ensure that a field is always unique in all collections in a
sharded cluster.
Shard GridFS Data Store (page 544) Choose whether to shard GridFS data in a sharded collection.
Create Chunks in a Sharded Cluster
Pre-splitting the chunk ranges in an empty sharded collection allows clients to insert data into an already partitioned
collection. In most situations a sharded cluster will create and distribute chunks automatically without user intervention. However, in a limited number of cases, MongoDB cannot create enough chunks or distribute data fast enough to
support required throughput. For example:
If you want to partition an existing data collection that resides on a single shard.
If you want to ingest a large volume of data into a cluster that isnt balanced, or where the ingestion of data will
lead to data imbalance. For example, monotonically increasing or decreasing shard keys insert all data into a
single chunk.
These operations are resource intensive for several reasons:
Chunk migration requires copying all the data in the chunk from one shard to another.
MongoDB can migrate only a single chunk at a time.
MongoDB creates splits only after an insert operation.
Warning: Only pre-split an empty collection. If a collection already has data, MongoDB automatically splits the
collections data when you enable sharding for the collection. Subsequent attempts to manually create splits can
lead to unpredictable chunk ranges and sizes as well as inefficient or ineffective balancing behavior.
To create chunks manually, use the following procedure:

536

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

1. Split empty chunks in your collection by manually performing the split command on chunks.
Example
To create chunks for documents in the myapp.users collection using the email field as the shard key, use
the following operation in the mongo shell:
for ( var x=97; x<97+26; x++ ){
for( var y=97; y<97+26; y+=6 ) {
var prefix = String.fromCharCode(x) + String.fromCharCode(y);
db.runCommand( { split : "myapp.users" , middle : { email : prefix } } );
}
}

This assumes a collection size of 100 million documents.

For information on the balancer and automatic distribution of chunks across shards, see Cluster Balancer
(page 501) and Chunk Migration (page 503). For information on manually migrating chunks, see Migrate
Chunks in a Sharded Cluster (page 538).
Split Chunks in a Sharded Cluster
Normally, MongoDB splits a chunk after an insert if the chunk exceeds the maximum chunk size (page 504). However,
you may want to split chunks manually if:
you have a large amount of data in your cluster and very few chunks, as is the case after deploying a cluster
using existing data.
you expect to add a large amount of data that would initially reside in a single chunk or shard. For example, you
plan to insert a large amount of data with shard key values between 300 and 400, but all values of your shard
keys are between 250 and 500 are in a single chunk.
Note: Chunks cannot be merged or combined once theyve been split.
The balancer may migrate recently split chunks to a new shard immediately if mongos predicts future insertions will
benefit from the move. The balancer does not distinguish between chunks split manually and those split automatically
by the system.
Warning: Be careful when splitting data in a sharded collection to create new chunks. When you shard a
collection that has existing data, MongoDB automatically creates chunks to evenly distribute the collection. To
split data effectively in a sharded cluster you must consider the number of documents in a chunk and the average
document size to create a uniform chunk size. When chunks have irregular sizes, shards may have an equal number
of chunks but have very different data sizes. Avoid creating splits that lead to a collection with differently sized
chunks.
Use sh.status() to determine the current chunk ranges across the cluster.
To split chunks manually, use the split command with either fields middle or find. The mongo shell provides
the helper methods sh.splitFind() and sh.splitAt().
splitFind() splits the chunk that contains the first document returned that matches this query into two equally
sized chunks. You must specify the full namespace (i.e. <database>.<collection>) of the sharded collection
to splitFind(). The query in splitFind() does not need to use the shard key, though it nearly always makes
sense to do so.
Example

9.3. Sharded Cluster Tutorials

537

MongoDB Documentation, Release 2.4.9

The following command splits the chunk that contains the value of 63109 for the zipcode field in the people
collection of the records database:
sh.splitFind( "records.people", { "zipcode": 63109 } )

Use splitAt() to split a chunk in two, using the queried document as the lower bound in the new chunk:
Example
The following command splits the chunk that contains the value of 63109 for the zipcode field in the people
collection of the records database.
sh.splitAt( "records.people", { "zipcode": 63109 } )

Note: splitAt() does not necessarily split the chunk into two equally sized chunks. The split occurs at the location
of the document matching the query, regardless of where that document is in the chunk.

Migrate Chunks in a Sharded Cluster

In most circumstances, you should let the automatic balancer migrate chunks between shards. However, you may
want to migrate chunks manually in a few cases:
When pre-splitting an empty collection, migrate chunks manually to distribute them evenly across the shards.
Use pre-splitting in limited situations to support bulk data ingestion.
If the balancer in an active cluster cannot distribute chunks within the balancing window (page 532), then you
will have to migrate chunks manually.
To manually migrate chunks, use the moveChunk command. For more information on how the automatic balancer
moves chunks between shards, see Cluster Balancer (page 501) and Chunk Migration (page 503).
Example
Migrate a single chunk
The following example assumes that the field username is the shard key for a collection named users in the
myapp database, and that the value smith exists within the chunk to migrate. Migrate the chunk using the following
command in the mongo shell.
db.adminCommand( { moveChunk : "myapp.users",
find : {username : "smith"},
to : "mongodb-shard3.example.net" } )

This command moves the chunk that includes the shard key value smith to the shard named
mongodb-shard3.example.net. The command will block until the migration is complete.
Tip
To return a list of shards, use the listShards command.

Example
Evenly migrate chunks
To evenly migrate chunks for the myapp.users collection, put each prefix chunk on the next shard from the other
and run the following commands in the mongo shell:

538

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

var shServer = [ "sh0.example.net", "sh1.example.net", "sh2.example.net", "sh3.example.net", "sh4.exa

for ( var x=97; x<97+26; x++ ){
for( var y=97; y<97+26; y+=6 ) {
var prefix = String.fromCharCode(x) + String.fromCharCode(y);
db.adminCommand({moveChunk : "myapp.users", find : {email : prefix}, to : shServer[(y-97)/6]})
}
}

See Create Chunks in a Sharded Cluster (page 536) for an introduction to pre-splitting.
New in version 2.2: The moveChunk command has the: _secondaryThrottle parameter. When set to true,
MongoDB ensures that changes to shards as part of chunk migrations replicate to secondaries throughout the migration
operation. For more information, see Require Replication before Chunk Migration (Secondary Throttle) (page 530).
Changed in version 2.4: In 2.4, _secondaryThrottle is true by default.
Warning: The moveChunk command may produce the following error message:
The collection's metadata lock is already taken.

This occurs when clients have too many open cursors that access the migrating chunk. You may either wait until
the cursors complete their operations or close the cursors manually.

Modify Chunk Size in a Sharded Cluster

When the first mongos connects to a set of config servers, it initializes the sharded cluster with a default chunk size
of 64 megabytes. This default chunk size works well for most deployments; however, if you notice that automatic
migrations have more I/O than your hardware can handle, you may want to reduce the chunk size. For automatic splits
and migrations, a small chunk size leads to more rapid and frequent migrations.
To modify the chunk size, use the following procedure:
1. Connect to any mongos in the cluster using the mongo shell.
2. Issue the following command to switch to the Config Database (page 547):
use config

3. Issue the following save() operation to store the global chunk size configuration value:
db.settings.save( { _id:"chunksize", value: <size> } )

Note: The chunkSize and --chunkSize options, passed at runtime to the mongos, do not affect the chunk size
after you have initialized the cluster.
To avoid confusion, always set the chunk size using the above procedure instead of the runtime options.
Modifying the chunk size has several limitations:
Automatic splitting only occurs on insert or update.
If you lower the chunk size, it may take time for all chunks to split to the new size.
Splits cannot be undone.
If you increase the chunk size, existing chunks grow only through insertion or updates until they reach the new
size.

9.3. Sharded Cluster Tutorials

539

MongoDB Documentation, Release 2.4.9

Tag Aware Sharding

MongoDB supports tagging a range of shard key values to associate that range with a shard or group of shards. Those
shards receive all inserts within the tagged range.
The balancer obeys tagged range associations, which enables the following deployment patterns:
isolate a specific subset of data on a specific set of shards.
ensure that the most relevant data reside on shards that are geographically closest to the application servers.
This document describes the behavior, operation, and use of tag aware sharding in MongoDB deployments.
Considerations

Shard key range tags are distinct from replica set member tags (page 408).
Hash-based sharding does not support tag-aware sharding.
Behavior and Operations

The balancer migrates chunks of documents in a sharded collections to the shards associated with a tag that has a shard
key range with an upper bound greater than the chunks lower bound.
During balancing rounds, if the balancer detects that any chunks violate configured tags, the balancer migrates chunks
in tagged ranges to shards associated with those tags.
After configuring tags with a shard key range, and associating it with a shard or shards, the cluster may take some time
to balance the data among the shards. This depends on the division of chunks and the current distribution of data in
the cluster.
Once configured, the balancer respects tag ranges during future balancing rounds (page 501).
See also:
Manage Shard Tags (page 541)
Chunks that Span Multiple Tag Ranges

A single chunk may contain data with a shard key values that falls into ranges associated with more than one tag. To
accommodate these situations, the balancer may migrate chunks to shards that contain shard key values that exceed
the upper bound of the selected tag range.
Example
Given a sharded collection with two configured tag ranges:
Shard key values between 100 and 200 have tags to direct corresponding chunks to shards tagged NYC.
Shard key values between 200 and 300 have tags to direct corresponding chunks to shards tagged SFO.
For this collection cluster, the balancer will migrate a chunk with shard key values ranging between 150 and 220 to
a shard tagged NYC, since 150 is closer to 200 than 300.
To ensure that your collection has no potentially ambiguously tagged chunks, create splits on your tag boundaries
(page 537). You can then manually migrate chunks to the appropriate shards, or wait for the balancer to automatically
migrate these chunks.

540

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Manage Shard Tags

In a sharded cluster, you can use tags to associate specific ranges of a shard key with a specific shard or subset of
shards.
Tag a Shard

Associate tags with a particular shard using the sh.addShardTag() method when connected to a mongos instance. A single shard may have multiple tags, and multiple shards may also have the same tag.
Example
The following example adds the tag NYC to two shards, and the tags SFO and NRT to a third shard:
sh.addShardTag("shard0000",
sh.addShardTag("shard0001",
sh.addShardTag("shard0002",
sh.addShardTag("shard0002",

"NYC")
"NYC")
"SFO")
"NRT")

You may remove tags from a particular shard using the sh.removeShardTag() method when connected to a
mongos instance, as in the following example, which removes the NRT tag from a shard:
sh.removeShardTag("shard0002", "NRT")

Tag a Shard Key Range

To assign a tag to a range of shard keys use the sh.addTagRange() method when connected to a mongos instance.
Any given shard key range may only have one assigned tag. You cannot overlap defined ranges, or tag the same range
more than once.
Example
Given a collection named users in the records database, sharded by the zipcode field. The following operations
assign:
two ranges of zip codes in Manhattan and Brooklyn the NYC tag
one range of zip codes in San Francisco the SFO tag
sh.addTagRange("records.users", { zipcode: "10001" }, { zipcode: "10281" }, "NYC")
sh.addTagRange("records.users", { zipcode: "11201" }, { zipcode: "11240" }, "NYC")
sh.addTagRange("records.users", { zipcode: "94102" }, { zipcode: "94135" }, "SFO")

Note: Shard ranges are always inclusive of the lower value and exclusive of the upper boundary.

Remove a Tag From a Shard Key Range

The mongod does not provide a helper for removing a tag range. You may delete tag assignment from a shard key
range by removing the corresponding document from the tags (page 552) collection of the config database.
Each document in the tags (page 552) holds the namespace of the sharded collection and a minimum shard key
value.

9.3. Sharded Cluster Tutorials

541

MongoDB Documentation, Release 2.4.9

Example
The following example removes the NYC tag assignment for the range of zip codes within Manhattan:
use config
db.tags.remove({ _id: { ns: "records.users", min: { zipcode: "10001" }}, tag: "NYC" })

View Existing Shard Tags

The output from sh.status() lists tags associated with a shard, if any, for each shard. A shards tags exist in the
shards document in the shards (page 552) collection of the config database. To return all shards with a specific
tag, use a sequence of operations that resemble the following, which will return only those shards tagged with NYC:
use config
db.shards.find({ tags: "NYC" })

You can find tag ranges for all namespaces in the tags (page 552) collection of the config database. The output of
sh.status() displays all tag ranges. To return all shard key ranges tagged with NYC, use the following sequence
of operations:
use config
db.tags.find({ tags: "NYC" })

Enforce Unique Keys for Sharded Collections

Overview

The unique constraint on indexes ensures that only one document can have a value for a field in a collection. For
sharded collections these unique indexes cannot enforce uniqueness because insert and indexing operations are local
to each shard.
MongoDB does not support creating new unique indexes in sharded clusters and will not allow you to shard collections
with unique indexes on fields other than the _id field.
If you need to ensure that a field is always unique in all collections in a sharded environment, there are three options:
1. Enforce uniqueness of the shard key (page 492).
MongoDB can enforce uniqueness for the shard key. For compound shard keys, MongoDB will enforce uniqueness on the entire key combination, and not for a specific component of the shard key.
You cannot specify a unique constraint on a hashed index (page 333).
2. Use a secondary collection to enforce uniqueness.
Create a minimal collection that only contains the unique field and a reference to a document in the main
collection. If you always insert into a secondary collection before inserting to the main collection, MongoDB
will produce an error if you attempt to use a duplicate key.
If you have a small data set, you may not need to shard this collection and you can create multiple unique
indexes. Otherwise you can shard on a single unique key.
3. Use guaranteed unique identifiers.
Universally unique identifiers (i.e. UUID) like the ObjectId are guaranteed to be unique.

542

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Procedures

Unique Constraints on the Shard Key

Process To shard a collection using the unique constraint, specify the shardCollection command in the
following form:
db.runCommand( { shardCollection : "test.users" , key : { email : 1 } , unique : true } );

Remember that the _id field index is always unique. By default, MongoDB inserts an ObjectId into the _id field.
However, you can manually insert your own value into the _id field and use this as the shard key. To use the _id
field as the shard key, use the following operation:
db.runCommand( { shardCollection : "test.users" } )

Limitations
You can only enforce uniqueness on one single field in the collection using this method.
If you use a compound shard key, you can only enforce uniqueness on the combination of component keys in
the shard key.
In most cases, the best shard keys are compound keys that include elements that permit write scaling (page 493) and
query isolation (page 494), as well as high cardinality (page 512). These ideal shard keys are not often the same keys
that require uniqueness and enforcing unique values in these collections requires a different approach.
Unique Constraints on Arbitrary Fields If you cannot use a unique field as the shard key or if you need to enforce
uniqueness over multiple fields, you must create another collection to act as a proxy collection. This collection must
contain both a reference to the original document (i.e. its ObjectId) and the unique key.
If you must shard this proxy collection, then shard on the unique key using the above procedure (page 543); otherwise, you can simply create multiple unique indexes on the collection.
Process Consider the following for the proxy collection:
{
"_id" : ObjectId("...")
"email" ": "..."
}

The _id field holds the ObjectId of the document it reflects, and the email field is the field on which you want to
ensure uniqueness.
To shard this collection, use the following operation using the email field as the shard key:
db.runCommand( { shardCollection : "records.proxy" , key : { email : 1 } , unique : true } );

If you do not need to shard the proxy collection, use the following command to create a unique index on the email
field:
db.proxy.ensureIndex( { "email" : 1 }, {unique : true} )

You may create multiple unique indexes on this collection if you do not plan to shard the proxy collection.
To insert documents, use the following procedure in the JavaScript shell:

9.3. Sharded Cluster Tutorials

543

MongoDB Documentation, Release 2.4.9

use records;
var primary_id = ObjectId();
db.proxy.insert({
"_id" : primary_id
"email" : "[email protected]"
})
// if: the above operation returns successfully,
// then continue:
db.information.insert({
"_id" : primary_id
"email": "[email protected]"
// additional information...
})

You must insert a document into the proxy collection first. If this operation succeeds, the email field is unique, and
you may continue by inserting the actual document into the information collection.
See
The full documentation of: ensureIndex() and shardCollection.

Considerations
Your application must catch errors when inserting documents into the proxy collection and must enforce
consistency between the two collections.
If the proxy collection requires sharding, you must shard on the single field on which you want to enforce
uniqueness.
To enforce uniqueness on more than one field using sharded proxy collections, you must have one proxy collection for every field for which to enforce uniqueness. If you create multiple unique indexes on a single proxy
collection, you will not be able to shard proxy collections.
Use Guaranteed Unique Identifier The best way to ensure a field has unique values is to generate universally
unique identifiers (UUID,) such as MongoDBs ObjectId values.
This approach is particularly useful for the_id field, which must be unique: for collections where you are not
sharding by the _id field the application is responsible for ensuring that the _id field is unique.
Shard GridFS Data Store
When sharding a GridFS store, consider the following:
files Collection

Most deployments will not need to shard the files collection. The files collection is typically small, and only
contains metadata. None of the required keys for GridFS lend themselves to an even distribution in a sharded situation.
If you must shard the files collection, use the _id field possibly in combination with an application field.
Leaving files unsharded means that all the file metadata documents live on one shard. For production GridFS stores
you must store the files collection on a replica set.
544

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

chunks Collection

To shard the chunks collection by { files_id :

1 , n :

1 }, issue commands similar to the following:

db.fs.chunks.ensureIndex( { files_id : 1 , n : 1 } )
db.runCommand( { shardCollection : "test.fs.chunks" , key : { files_id : 1 , n : 1 } } )

You may also want to shard using just the file_id field, as in the following operation:
db.runCommand( { shardCollection : "test.fs.chunks" , key : {

Important: { files_id : 1 , n :
for the chunks collection of a GridFS store.

1 } and { files_id :

files_id : 1 } } )

1 } are the only supported shard keys

Note: Changed in version 2.2.

Before 2.2, you had to create an additional index on files_id to shard using only this field.
The default files_id value is an ObjectId, as a result the values of files_id are always ascending, and applications will insert all new GridFS data to a single chunk and shard. If your write load is too high for a single server to
handle, consider a different shard key or use a different value for _id in the files collection.

9.3.4 Troubleshoot Sharded Clusters

This section describes common strategies for troubleshooting sharded cluster deployments.
Config Database String Error
Start all mongos instances in a sharded cluster with an identical configdb string. If a mongos instance tries
to connect to the sharded cluster with a configdb string that does not exactly match the string used by the other
mongos instances, including the order of the hosts, the following errors occur:
could not initialize sharding on connection

And:
mongos specified a different config database string

To solve the issue, restart the mongos with the correct string.
Cursor Fails Because of Stale Config Data
A query returns the following warning when one or more of the mongos instances has not yet updated its cache of
the clusters metadata from the config database:
could not initialize cursor across all shards because : stale config detected

This warning should not propagate back to your application. The warning will repeat until all the mongos instances
refresh their caches. To force an instance to refresh its cache, run the flushRouterConfig command.

9.3. Sharded Cluster Tutorials

545

MongoDB Documentation, Release 2.4.9

Avoid Downtime when Moving Config Servers

Use CNAMEs to identify your config servers to the cluster so that you can rename and renumber your config servers
without downtime.

9.4 Sharding Reference

9.4.1 Sharding Methods in the mongo Shell
Name
Description
sh._adminCommand Runs a database command against the admin database, like db.runCommand(), but
can confirm that it is issued against a mongos.
sh._checkFullName()
Tests a namespace to determine if its well formed.
sh._checkMongos() Tests to see if the mongo shell is connected to a mongos instance.
sh._lastMigration()
Reports on the last chunk migration.
sh.addShard()
Adds a shard to a sharded cluster.
sh.addShardTag() Associates a shard with a tag, to support tag aware sharding (page 540).
sh.addTagRange() Associates range of shard keys with a shard tag, to support tag aware sharding
(page 540).
sh.disableBalancing()
Disable balancing on a single collection in a sharded database. Does not affect
balancing of other collections in a sharded cluster.
sh.enableBalancing()
Activates the sharded collection balancer process if previously disabled using
sh.disableBalancing().
sh.enableSharding()
Enables sharding on a specific database.
sh.getBalancerHost()
Returns the name of a mongos thats responsible for the balancer process.
sh.getBalancerState()
Returns a boolean to report if the balancer is currently enabled.
sh.help()
Returns help text for the sh methods.
sh.isBalancerRunning()
Returns a boolean to report if the balancer process is currently migrating chunks.
sh.moveChunk()
Migrates a chunk in a sharded cluster.
sh.removeShardTag()
Removes the association between a shard and a shard tag shard tag.
sh.setBalancerState()
Enables or disables the balancer which migrates chunks between shards.
sh.shardCollection()
Enables sharding for a collection.
sh.splitAt()
Divides an existing chunk into two chunks using a specific value of the shard key as the
dividing point.
sh.splitFind()
Divides an existing chunk that contains a document matching a query into two
approximately equal chunks.
sh.startBalancer()Enables the balancer and waits for balancing to start.
sh.status()
Reports on the status of a sharded cluster, as db.printShardingStatus().
sh.stopBalancer() Disables the balancer and waits for any in progress balancing rounds to complete.
sh.waitForBalancer()
Internal. Waits for the balancer state to change.
sh.waitForBalancerOff()
Internal. Waits until the balancer stops running.
sh.waitForDLock() Internal. Waits for a specified distributed sharded cluster lock.
sh.waitForPingChange()
Internal. Waits for a change in ping state from one of the mongos in the sharded cluster.

9.4.2 Sharding Database Commands

The following database commands support sharded clusters.

546

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Name
Description
flushRouterConfig Forces an update to the cluster metadata cached by a mongos.
addShard
Adds a shard to a sharded cluster.
checkShardingIndexInternal command that validates index on shard key.
enableSharding
Enables sharding on a specific database.
listShards
Returns a list of configured shards.
removeShard
Starts the process of removing a shard from a sharded cluster.
getShardMap
Internal command that reports on the state of a sharded cluster.
getShardVersion
Internal command that returns the config server version.
setShardVersion
Internal command to sets the config server version.
shardCollection
Enables the sharding functionality for a collection, allowing the collection to be
sharded.
shardingState
Reports whether the mongod is a member of a sharded cluster.
unsetSharding
Internal command that affects connections between instances in a MongoDB
deployment.
split
Creates a new chunk.
splitChunk
Internal command to split chunk. Instead use the methods sh.splitFind() and
sh.splitAt().
splitVector
Internal command that determines split points.
medianKey
Deprecated internal command. See splitVector.
moveChunk
Internal command that migrates chunks between shards.
movePrimary
Reassigns the primary shard when removing a shard from a sharded cluster.
isdbgrid
Verifies that a process is a mongos.

9.4.3 Reference Documentation

Config Database (page 547) Complete documentation of the content of the local database that MongoDB uses to
store sharded cluster metadata.
Config Database
The config database supports sharded cluster operation. See the Sharding (page 479) section of this manual for full
documentation of sharded clusters.
Important: Consider the schema of the config database internal and may change between releases of MongoDB.
The config database is not a dependable API, and users should not write data to the config database in the course
of normal operation or maintenance.
Warning: Modification of the config database on a functioning system may lead to instability or inconsistent
data sets. If you must modify the config database, use mongodump to create a full backup of the config
database.
To access the config database, connect to a mongos instance in a sharded cluster, and use the following helper:
use config

You can return a list of the collections, with the following helper:
show collections

9.4. Sharding Reference

547

MongoDB Documentation, Release 2.4.9

Collections

config
config.changelog

Internal MongoDB Metadata

The config (page 548) database is internal: applications and administrators should not modify or depend upon
its content in the course of normal operation.
The changelog (page 548) collection stores a document for each change to the metadata of a sharded collection.
Example
The following example displays a single record of a chunk split from a changelog (page 548) collection:
{
"_id" : "<hostname>-<timestamp>-<increment>",
"server" : "<hostname><:port>",
"clientAddr" : "127.0.0.1:63381",
"time" : ISODate("2012-12-11T14:09:21.039Z"),
"what" : "split",
"ns" : "<database>.<collection>",
"details" : {
"before" : {
"min" : {
"<database>" : { $minKey : 1 }
},
"max" : {
"<database>" : { $maxKey : 1 }
},
"lastmod" : Timestamp(1000, 0),
"lastmodEpoch" : ObjectId("000000000000000000000000")
},
"left" : {
"min" : {
"<database>" : { $minKey : 1 }
},
"max" : {
"<database>" : "<value>"
},
"lastmod" : Timestamp(1000, 1),
"lastmodEpoch" : ObjectId(<...>)
},
"right" : {
"min" : {
"<database>" : "<value>"
},
"max" : {
"<database>" : { $maxKey : 1 }
},
"lastmod" : Timestamp(1000, 2),
"lastmodEpoch" : ObjectId("<...>")
}
}
}

548

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Each document in the changelog (page 548) collection contains the following fields:
config.changelog._id
The value of changelog._id is: <hostname>-<timestamp>-<increment>.
config.changelog.server
The hostname of the server that holds this data.
config.changelog.clientAddr
A string that holds the address of the client, a mongos instance that initiates this change.
config.changelog.time
A ISODate timestamp that reflects when the change occurred.
config.changelog.what
Reflects the type of change recorded. Possible values are:
dropCollection
dropCollection.start
dropDatabase
dropDatabase.start
moveChunk.start
moveChunk.commit
split
multi-split
config.changelog.ns
Namespace where the change occurred.
config.changelog.details
A document that contains additional details regarding the change.
(page 549) document depends on the type of change.

The structure of the details

config.chunks

Internal MongoDB Metadata

The config (page 548) database is internal: applications and administrators should not modify or depend upon
its content in the course of normal operation.
The chunks (page 549) collection stores a document for each chunk in the cluster. Consider the following
example of a document for a chunk named records.pets-animal_\"cat\":
{
"_id" : "mydb.foo-a_\"cat\"",
"lastmod" : Timestamp(1000, 3),
"lastmodEpoch" : ObjectId("5078407bd58b175c5c225fdc"),
"ns" : "mydb.foo",
"min" : {
"animal" : "cat"
},
"max" : {
"animal" : "dog"
},

9.4. Sharding Reference

549

MongoDB Documentation, Release 2.4.9

"shard" : "shard0004"
}

These documents store the range of values for the shard key that describe the chunk in the min and max fields.
Additionally the shard field identifies the shard in the cluster that owns the chunk.
config.collections

Internal MongoDB Metadata

The config (page 548) database is internal: applications and administrators should not modify or depend upon
its content in the course of normal operation.
The collections (page 550) collection stores a document for each sharded collection in the cluster. Given
a collection named pets in the records database, a document in the collections (page 550) collection
would resemble the following:
{
"_id" : "records.pets",
"lastmod" : ISODate("1970-01-16T15:00:58.107Z"),
"dropped" : false,
"key" : {
"a" : 1
},
"unique" : false,
"lastmodEpoch" : ObjectId("5078407bd58b175c5c225fdc")
}

config.databases

Internal MongoDB Metadata

The config (page 548) database is internal: applications and administrators should not modify or depend upon
its content in the course of normal operation.
The databases (page 550) collection stores a document for each database in the cluster, and tracks if the
database has sharding enabled. databases (page 550) represents each database in a distinct document. When
a databases have sharding enabled, the primary field holds the name of the primary shard.
{ "_id" : "admin", "partitioned" : false, "primary" : "config" }
{ "_id" : "mydb", "partitioned" : true, "primary" : "shard0000" }

config.lockpings

Internal MongoDB Metadata

The config (page 548) database is internal: applications and administrators should not modify or depend upon
its content in the course of normal operation.
The lockpings (page 550) collection keeps track of the active components in the sharded cluster. Given
a cluster with a mongos running on example.com:30000, the document in the lockpings (page 550)
collection would resemble:
{ "_id" : "example.com:30000:1350047994:16807", "ping" : ISODate("2012-10-12T18:32:54.892Z") }

550

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

config.locks

Internal MongoDB Metadata

The config (page 548) database is internal: applications and administrators should not modify or depend upon
its content in the course of normal operation.
The locks (page 550) collection stores a distributed lock. This ensures that only one mongos instance can
perform administrative tasks on the cluster at once. The mongos acting as balancer takes a lock by inserting a
document resembling the following into the locks collection.
{
"_id" : "balancer",
"process" : "example.net:40000:1350402818:16807",
"state" : 2,
"ts" : ObjectId("507daeedf40e1879df62e5f3"),
"when" : ISODate("2012-10-16T19:01:01.593Z"),
"who" : "example.net:40000:1350402818:16807:Balancer:282475249",
"why" : "doing balance round"
}

If a mongos holds the balancer lock, the state field has a value of 2, which means that balancer is active.
The when field indicates when the balancer began the current operation.
Changed in version 2.0: The value of the state field was 1 before MongoDB 2.0.
config.mongos

Internal MongoDB Metadata

The config (page 548) database is internal: applications and administrators should not modify or depend upon
its content in the course of normal operation.
The mongos (page 551) collection stores a document for each mongos instance affiliated with the cluster.
mongos instances send pings to all members of the cluster every 30 seconds so the cluster can verify that the
mongos is active. The ping field shows the time of the last ping, while the up field reports the uptime of the
mongos as of the last ping. The cluster maintains this collection for reporting purposes.
The following document shows the status of the mongos running on example.com:30000.

{ "_id" : "example.com:30000", "ping" : ISODate("2012-10-12T17:08:13.538Z"), "up" : 13699, "wait

config.settings

Internal MongoDB Metadata

The config (page 548) database is internal: applications and administrators should not modify or depend upon
its content in the course of normal operation.
The settings (page 551) collection holds the following sharding configuration settings:
Chunk size. To change chunk size, see Modify Chunk Size in a Sharded Cluster (page 539).
Balancer status. To change status, see Disable the Balancer (page 533).
The following is an example settings collection:

9.4. Sharding Reference

551

MongoDB Documentation, Release 2.4.9

{ "_id" : "chunksize", "value" : 64 }

{ "_id" : "balancer", "stopped" : false }

config.shards

Internal MongoDB Metadata

The config (page 548) database is internal: applications and administrators should not modify or depend upon
its content in the course of normal operation.
The shards (page 552) collection represents each shard in the cluster in a separate document. If the shard
is a replica set, the host field displays the name of the replica set, then a slash, then the hostname, as in the
following example:
{ "_id" : "shard0000", "host" : "shard1/localhost:30000" }

If the shard has tags (page 540) assigned, this document has a tags field, that holds an array of the tags, as in
the following example:
{ "_id" : "shard0001", "host" : "localhost:30001", "tags": [ "NYC" ] }

config.tags

Internal MongoDB Metadata

The config (page 548) database is internal: applications and administrators should not modify or depend upon
its content in the course of normal operation.
The tags (page 552) collection holds documents for each tagged shard key range in the cluster. The documents
in the tags (page 552) collection resemble the following:
{
"_id" : { "ns" : "records.users", "min" : { "zipcode" : "10001" } },
"ns" : "records.users",
"min" : { "zipcode" : "10001" },
"max" : { "zipcode" : "10281" },
"tag" : "NYC"
}

config.version

Internal MongoDB Metadata

The config (page 548) database is internal: applications and administrators should not modify or depend upon
its content in the course of normal operation.
The version (page 552) collection holds the current metadata version number. This collection contains only
one document:
To access the version (page 552) collection you must use the db.getCollection() method. For example, to display the collections document:
mongos> db.getCollection("version").find()
{ "_id" : 1, "version" : 3 }

552

Chapter 9. Sharding

MongoDB Documentation, Release 2.4.9

Note: Like all databases in MongoDB, the config database contains a system.indexes (page 227) collection
contains metadata for all indexes in the database for information on indexes, see Indexes (page 313).

9.4. Sharding Reference

553

MongoDB Documentation, Release 2.4.9

554

Chapter 9. Sharding

CHAPTER 10

Frequently Asked Questions

10.1 FAQ: MongoDB Fundamentals

This document addresses basic high level questions about MongoDB and its use.
If you dont find the answer youre looking for, check the complete list of FAQs (page 555) or post your question to
the MongoDB User Mailing List1 .

10.1.1 What kind of database is MongoDB?

MongoDB is a document-oriented DBMS. Think of MySQL but with JSON-like objects comprising the data model,
rather than RDBMS tables. Significantly, MongoDB supports neither joins nor transactions. However, it features
secondary indexes, an expressive query language, atomic writes on a per-document level, and fully-consistent reads.
Operationally, MongoDB features master-slave replication with automated failover and built-in horizontal scaling via
automated range-based partitioning.
Note: MongoDB uses BSON, a binary object format similar to, but more expressive than JSON.

10.1.2 Do MongoDB databases have tables?

Instead of tables, a MongoDB database stores its data in collections, which are the rough equivalent of RDBMS tables.
A collection holds one or more documents, which corresponds to a record or a row in a relational database table, and
each document has one or more fields, which corresponds to a column in a relational database table.
Collections have important differences from RDBMS tables. Documents in a single collection may have a unique
combination and set of fields. Documents need not have identical fields. You can add a field to some documents in a
collection without adding that field to all documents in the collection.
See
SQL to MongoDB Mapping Chart (page 85)
1 https://groups.google.com/forum/?fromgroups#!forum/mongodb-user

555

MongoDB Documentation, Release 2.4.9

10.1.3 Do MongoDB databases have schemas?

MongoDB uses dynamic schemas. You can create collections without defining the structure, i.e. the fields or the types
of their values, of the documents in the collection. You can change the structure of documents simply by adding new
fields or deleting existing ones. Documents in a collection need not have an identical set of fields.
In practice, it is common for the documents in a collection to have a largely homogeneous structure; however, this
is not a requirement. MongoDBs flexible schemas mean that schema migration and augmentation are very easy in
practice, and you will rarely, if ever, need to write scripts that perform alter table type operations, which simplifies
and facilitates iterative software development with MongoDB.
See
SQL to MongoDB Mapping Chart (page 85)

10.1.4 What languages can I use to work with MongoDB?

MongoDB client drivers exist for all of the most popular programming languages, and many other ones. See the latest
list of drivers2 for details.
See also:
MongoDB Drivers and Client Libraries (page 95).

10.1.5 Does MongoDB support SQL?

No.
However, MongoDB does support a rich, ad-hoc query language of its own.
See also:
http://docs.mongodb.org/manualreference/operator

10.1.6 What are typical uses for MongoDB?

MongoDB has a general-purpose design, making it appropriate for a large number of use cases. Examples include
content management systems, mobile applications, gaming, e-commerce, analytics, archiving, and logging.
Do not use MongoDB for systems that require SQL, joins, and multi-object transactions.

10.1.7 Does MongoDB support ACID transactions?

MongoDB does not support multi-document transactions.
However, MongoDB does provide atomic operations on a single document. Often these document-level atomic operations are sufficient to solve problems that would require ACID transactions in a relational database.
For example, in MongoDB, you can embed related data in nested arrays or nested documents within a single document
and update the entire document in a single atomic operation. Relational databases might represent the same kind of
data with multiple tables and rows, which would require transaction support to update the data atomically.
2 http://docs.mongodb.org/ecosystem/drivers

556

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

10.1.8 Does MongoDB require a lot of RAM?

Not necessarily. Its certainly possible to run MongoDB on a machine with a small amount of free RAM.
MongoDB automatically uses all free memory on the machine as its cache. System resource monitors show that
MongoDB uses a lot of memory, but its usage is dynamic. If another process suddenly needs half the servers RAM,
MongoDB will yield cached memory to the other process.
Technically, the operating systems virtual memory subsystem manages MongoDBs memory. This means that MongoDB will use as much free memory as it can, swapping to disk as needed. Deployments with enough memory to fit
the applications working data set in RAM will achieve the best performance.
See also:
FAQ: MongoDB Diagnostics (page 587) for answers to additional questions about MongoDB and Memory use.

10.1.9 How do I configure the cache size?

MongoDB has no configurable cache. MongoDB uses all free memory on the system automatically by way of memorymapped files. Operating systems use the same approach with their file system caches.

10.1.10 Does MongoDB require a separate caching layer for application-level

caching?
No. In MongoDB, a documents representation in the database is similar to its representation in application memory.
This means the database already stores the usable form of data, making the data usable in both the persistent store and
in the application cache. This eliminates the need for a separate caching layer in the application.
This differs from relational databases, where caching data is more expensive. Relational databases must transform
data into object representations that applications can read and must store the transformed data in a separate cache: if
these transformation from data to application objects require joins, this process increases the overhead related to using
the database which increases the importance of the caching layer.

10.1.11 Does MongoDB handle caching?

Yes. MongoDB keeps all of the most recently used data in RAM. If you have created indexes for your queries and
your working data set fits in RAM, MongoDB serves all queries from memory.
MongoDB does not implement a query cache: MongoDB serves all queries directly from the indexes and/or data files.

10.1.12 Are writes written to disk immediately, or lazily?

Writes are physically written to the journal (page 234) within 100 milliseconds, by default. At that point, the write is
durable in the sense that after a pull-plug-from-wall event, the data will still be recoverable after a hard restart. See
journalCommitInterval for more information on the journal commit window.
While the journal commit is nearly instant, MongoDB writes to the data files lazily. MongoDB may wait to write
data to the data files for as much as one minute by default. This does not affect durability, as the journal has enough
information to ensure crash recovery. To change the interval for writing to the data files, see syncdelay.

10.1. FAQ: MongoDB Fundamentals

557

MongoDB Documentation, Release 2.4.9

10.1.13 What language is MongoDB written in?

MongoDB is implemented in C++. Drivers and client libraries are typically written in their respective languages,
although some drivers use C extensions for better performance.

10.1.14 What are the limitations of 32-bit versions of MongoDB?

MongoDB uses memory-mapped files (page 581). When running a 32-bit build of MongoDB, the total storage size
for the server, including data and indexes, is 2 gigabytes. For this reason, do not deploy MongoDB to production on
32-bit machines.
If youre running a 64-bit build of MongoDB, theres virtually no limit to storage size. For production deployments,
64-bit builds and operating systems are strongly recommended.
See also:
Blog Post: 32-bit Limitations3
Note: 32-bit builds disable journaling by default because journaling further limits the maximum amount of data that
the database can

10.2 FAQ: MongoDB for Application Developers

This document answers common questions about application development using MongoDB.
If you dont find the answer youre looking for, check the complete list of FAQs (page 555) or post your question to
the MongoDB User Mailing List4 .

10.2.1 What is a namespace in MongoDB?

A namespace is the concatenation of the database name and the collection names
between.

with a period character in

Collections are containers for documents that share one or more indexes. Databases are groups of collections stored
on disk using a single set of data files. 6
For an example acme.users namespace, acme is the database name and users is the collection name. Period
characters can occur in collection names, so that acme.user.history is a valid namespace, with acme as the
database name, and user.history as the collection name.
While data models like this appear to support nested collections, the collection namespace is flat, and there is no
difference from the perspective of MongoDB between acme, acme.users, and acme.records.

10.2.2 How do you copy all objects from one collection to another?
In the mongo shell, you can use the following operation to duplicate the entire collection:
db.source.copyTo(newCollection)
3 http://blog.mongodb.org/post/137788967/32-bit-limitations
4 https://groups.google.com/forum/?fromgroups#!forum/mongodb-user
5
6

Each index also has its own namespace.

MongoDB database have a configurable limit on the number of namespaces in a database.

558

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

Warning: When using db.collection.copyTo() check field types to ensure that the operation does
not remove type information from documents during the translation from BSON to JSON. Consider using
cloneCollection() to maintain type fidelity.
Also consider the cloneCollection command that may provide some of this functionality.

10.2.3 If you remove a document, does MongoDB remove it from disk?

Yes.
When you use remove(), the object will no longer exist in MongoDBs on-disk data storage.

10.2.4 When does MongoDB write updates to disk?

MongoDB flushes writes to disk on a regular interval. In the default configuration, MongoDB writes data to the
main data files on disk every 60 seconds and commits the journal roughly every 100 milliseconds. These values are
configurable with the journalCommitInterval and syncdelay.
These values represent the maximum amount of time between the completion of a write operation and the point when
the write is durable in the journal, if enabled, and when MongoDB flushes data to the disk. In many cases MongoDB
and the operating system flush data to disk more frequently, so that the above values represents a theoretical maximum.
However, by default, MongoDB uses a lazy strategy to write to disk. This is advantageous in situations where the
database receives a thousand increments to an object within one second, MongoDB only needs to flush this data to
disk once. In addition to the aforementioned configuration options, you can also use fsync and getLastError to
modify this strategy.

10.2.5 How do I do transactions and locking in MongoDB?

MongoDB does not have support for traditional locking or complex transactions with rollback. MongoDB aims to be
lightweight, fast, and predictable in its performance. This is similar to the MySQL MyISAM autocommit model. By
keeping transaction support extremely simple, MongoDB can provide greater performance especially for partitioned
or replicated systems with a number of database server processes.
MongoDB does have support for atomic operations within a single document. Given the possibilities provided by
nested documents, this feature provides support for a large number of use-cases.
See also:
The Isolate Sequence of Operations (page 76) page.

10.2.6 How do you aggregate data with MongoDB?

In version 2.1 and later, you can use the new aggregation framework (page 281), with the aggregate command.
MongoDB also supports map-reduce with the mapReduce command, as well as basic aggregation with the group,
count, and distinct. commands.
See also:
The Aggregation (page 277) page.

10.2. FAQ: MongoDB for Application Developers

559

MongoDB Documentation, Release 2.4.9

10.2.7 Why does MongoDB log so many Connection Accepted events?

If you see a very large number connection and re-connection messages in your MongoDB log, then clients are frequently connecting and disconnecting to the MongoDB server. This is normal behavior for applications that do not use
request pooling, such as CGI. Consider using FastCGI, an Apache Module, or some other kind of persistent application
server to decrease the connection overhead.
If these connections do not impact your performance you can use the run-time quiet option or the command-line
option --quiet to suppress these messages from the log.

10.2.8 Does MongoDB run on Amazon EBS?

Yes.
MongoDB users of all sizes have had a great deal of success using MongoDB on the EC2 platform using EBS disks.
See also:
Amazon EC27

10.2.9 Why are MongoDBs data files so large?

MongoDB aggressively preallocates data files to reserve space and avoid file system fragmentation. You can use the
smallfiles setting to modify the file preallocation strategy.
See also:
Why are the files in my data directory larger than the data in my database? (page 582)

10.2.10 How do I optimize storage use for small documents?

Each MongoDB document contains a certain amount of overhead. This overhead is normally insignificant but becomes
significant if all documents are just a few bytes, as might be the case if the documents in your collection only have one
or two fields.
Consider the following suggestions and strategies for optimizing storage utilization for these collections:
Use the _id field explicitly.
MongoDB clients automatically add an _id field to each document and generate a unique 12-byte ObjectId for
the _id field. Furthermore, MongoDB always indexes the _id field. For smaller documents this may account
for a significant amount of space.
To optimize storage use, users can specify a value for the _id field explicitly when inserting documents into the
collection. This strategy allows applications to store a value in the _id field that would have occupied space in
another portion of the document.
You can store any value in the _id field, but because this value serves as a primary key for documents in the
collection, it must uniquely identify them. If the fields value is not unique, then it cannot serve as a primary key
as there would be collisions in the collection.
Use shorter field names.
MongoDB stores all field names in every document. For most documents, this represents a small fraction of the
space used by a document; however, for small documents the field names may represent a proportionally large
amount of space. Consider a collection of documents that resemble the following:
7 http://docs.mongodb.org/ecosystem/platforms/amazon-ec2

560

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

{ last_name : "Smith", best_score: 3.9 }

If you shorten the filed named last_name to lname and the field name best_score to score, as follows,
you could save 9 bytes per document.
{ lname : "Smith", score : 3.9 }

Shortening field names reduces expressiveness and does not provide considerable benefit on for larger documents and where document overhead is not significant concern. Shorter field names do not reduce the size of
indexes, because indexes have a predefined structure.
In general it is not necessary to use short field names.
Embed documents.
In some cases you may want to embed documents in other documents and save on the per-document overhead.

10.2.11 When should I use GridFS?

For documents in a MongoDB collection, you should always use GridFS for storing files larger than 16 MB.
In some situations, storing large files may be more efficient in a MongoDB database than on a system-level filesystem.
If your filesystem limits the number of files in a directory, you can use GridFS to store as many files as needed.
When you want to keep your files and metadata automatically synced and deployed across a number of systems
and facilities. When using geographically distributed replica sets (page 396) MongoDB can distribute files and
their metadata automatically to a number of mongod instances and facilities.
When you want to access information from portions of large files without having to load whole files into memory,
you can use GridFS to recall sections of files without reading the entire file into memory.
Do not use GridFS if you need to update the content of the entire file atomically. As an alternative you can store
multiple versions of each file and specify the current version of the file in the metadata. You can update the metadata
field that indicates latest status in an atomic update after uploading the new version of the file, and later remove
previous versions if needed.
Furthermore, if your files are all smaller the 16 MB BSON Document Size limit, consider storing the file manually
within a single document. You may use the BinData data type to store the binary data. See your drivers (page 95)
documentation for details on using BinData.
For more information on GridFS, see GridFS (page 104).

10.2.12 How does MongoDB address SQL or Query injection?

BSON
As a client program assembles a query in MongoDB, it builds a BSON object, not a string. Thus traditional SQL
injection attacks are not a problem. More details and some nuances are covered below.
MongoDB represents queries as BSON objects. Typically client libraries (page 95) provide a convenient, injection
free, process to build these objects. Consider the following C++ example:
BSONObj my_query = BSON( "name" << a_name );
auto_ptr<DBClientCursor> cursor = c.query("tutorial.persons", my_query);

Here, my_query then will have a value such as { name : "Joe" }. If my_query contained special characters, for example ,, :, and {, the query simply wouldnt match any documents. For example, users cannot hijack a
query and convert it to a delete.
10.2. FAQ: MongoDB for Application Developers

561

MongoDB Documentation, Release 2.4.9

JavaScript
Note: You can disable all server-side execution of JavaScript, by passing the --noscripting option on the
command line or setting noscripting in a configuration file.
All of the following MongoDB operations permit you to run arbitrary JavaScript expressions directly on the server:
$where
db.eval()
mapReduce
group
You must exercise care in these cases to prevent users from submitting malicious JavaScript.
Fortunately, you can express most queries in MongoDB without JavaScript and for queries that require JavaScript, you
can mix JavaScript and non-JavaScript in a single query. Place all the user-supplied fields directly in a BSON field and
pass JavaScript code to the $where field.
If you need to pass user-supplied values in a $where clause, you may escape these values with the
CodeWScope mechanism. When you set user-submitted values as variables in the scope document, you can
avoid evaluating them on the database server.
If you need to use db.eval() with user supplied values, you can either use a CodeWScope or you can supply
extra arguments to your function. For instance:
db.eval(function(userVal){...},
user_value);

This will ensure that your application sends user_value to the database server as data rather than code.
Dollar Sign Operator Escaping
Field names in MongoDBs query language have semantic meaning. The dollar sign (i.e $) is a reserved character used
to represent operators (i.e. $inc.) Thus, you should ensure that your applications users cannot inject operators
into their inputs.
In some cases, you may wish to build a BSON object with a user-provided key. In these situations, keys will need
to substitute the reserved $ and . characters. Any character is sufficient, but consider using the Unicode full width
equivalents: U+FF04 (i.e. $) and U+FF0E (i.e. .).
Consider the following example:
BSONObj my_object = BSON( a_key << a_name );

The user may have supplied a $ value in the a_key value. At the same time, my_object might be { $where :
"things" }. Consider the following cases:
Insert. Inserting this into the database does no harm. The insert process does not evaluate the object as a query.
Note: MongoDB client drivers, if properly implemented, check for reserved characters in keys on inserts.
Update. The update() operation permits $ operators in the update argument but does not support the
$where operator. Still, some users may be able to inject operators that can manipulate a single document
only. Therefore your application should escape keys, as mentioned above, if reserved characters are possible.

562

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

Query Generally this is not a problem for queries that resemble { x : user_obj }: dollar signs are
not top level and have no effect. Theoretically it may be possible for the user to build a query themselves.
But checking the user-submitted content for $ characters in key names may help protect against this kind of
injection.
Driver-Specific Issues
See the PHP MongoDB Driver Security Notes8 page in the PHP driver documentation for more information

10.2.13 How does MongoDB provide concurrency?

MongoDB implements a readers-writer lock. This means that at any one time, only one client may be writing or any
number of clients may be reading, but that reading and writing cannot occur simultaneously.
In standalone and replica sets the locks scope applies to a single mongod instance or primary instance. In a sharded
cluster, locks apply to each individual shard, not to the whole cluster.
For more information, see FAQ: Concurrency (page 569).

10.2.14 What is the compare order for BSON types?

MongoDB permits documents within a single collection to have fields with different BSON types. For instance, the
following documents may exist within a single collection.
{ x: "string" }
{ x: 42 }

When comparing values of different BSON types, MongoDB uses the following comparison order, from lowest to
highest:
1. MinKey (internal type)
2. Null
3. Numbers (ints, longs, doubles)
4. Symbol, String
5. Object
6. Array
7. BinData
8. ObjectID
9. Boolean
10. Date, Timestamp
11. Regular Expression
12. MaxKey (internal type)
MongoDB treats some types as equivalent for comparison purposes. For instance, numeric types undergo conversion
before comparison.
The comparison treats a non-existent field as it would an empty BSON Object. As such, a sort on the a field in
documents { } and { a: null } would treat the documents as equivalent in sort order.
8 http://us.php.net/manual/en/mongo.security.php

10.2. FAQ: MongoDB for Application Developers

563

MongoDB Documentation, Release 2.4.9

With arrays, a less-than comparison or an ascending sort compares the smallest element of arrays, and a greater-than
comparison or a descending sort compares the largest element of the arrays. As such, when comparing a field whose
value is a single-element array (e.g. [ 1 ]) with non-array fields (e.g. 2), the comparison is between 1 and 2. A
comparison of an empty array (e.g. [ ]) treats the empty array as less than null or a missing field.
Consider the following mongo example:
db.test.insert(
db.test.insert(
db.test.insert(
db.test.insert(

{x
{x
{x
{x

:
:
:
:

3 } );
2.9 } );
new Date() } );
true } );

db.test.find().sort({x:1});
{ "_id" : ObjectId("4b03155dce8de6586fb002c7"),
{ "_id" : ObjectId("4b03154cce8de6586fb002c6"),
{ "_id" : ObjectId("4b031566ce8de6586fb002c9"),
{ "_id" : ObjectId("4b031563ce8de6586fb002c8"),

"x"
"x"
"x"
"x"

:
:
:
:

2.9 }
3 }
true }
"Tue Nov 17 2009 16:28:03 GMT-0500 (EST)" }

The $type operator provides access to BSON type comparison in the MongoDB query syntax. See the documentation
on BSON types and the $type operator for additional information.
Warning: Storing values of the different types in the same field in a collection is strongly discouraged.
See also:
The Tailable Cursors (page 74) page for an example of a C++ use of MinKey.

10.2.15 How do I query for fields that have null values?

Fields in a document may store null values, as in a notional collection, test, with the following documents:
{ _id: 1, cancelDate: null }
{ _id: 2 }

Different query operators treat null values differently:

The { cancelDate : null } query matches documents that either contains the cancelDate field
whose value is null or that do not contain the cancelDate field:
db.test.find( { cancelDate: null } )

The query returns both documents:

{ "_id" : 1, "cancelDate" : null }
{ "_id" : 2 }

The { cancelDate : { $type: 10 } } query matches documents that contains the cancelDate
field whose value is null only; i.e. the value of the cancelDate field is of BSON Type Null (i.e. 10) :
db.test.find( { cancelDate : { $type: 10 } } )

The query returns only the document that contains the null value:
{ "_id" : 1, "cancelDate" : null }

The { cancelDate :
cancelDate field:

564

{ $exists:

false } } query matches documents that do not contain the

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

db.test.find( { cancelDate : { $exists: false } } )

The query returns only the document that does not contain the cancelDate field:
{ "_id" : 2 }

See also:
The reference documentation for the $type and $exists operators.

10.2.16 Are there any restrictions on the names of Collections?

Collection names can be any UTF-8 string with the following exceptions:
A collection name should begin with a letter or an underscore.
The empty string ("") is not a valid collection name.
Collection names cannot contain the $ character. (version 2.2 only)
Collection names cannot contain the null character: \0
Do not name a collection using the system. prefix. MongoDB reserves system. for system collections,
such as the system.indexes collection.
The maximum size of a collection name is 128 characters, including the name of the database. However, for
maximum flexibility, collections should have names less than 80 characters.
If your collection name includes special characters, such as the underscore character, then to access the collection use
the db.getCollection() method or a similar method for your driver9 .
Example
To create a collection _foo and insert the { a :

1 } document, use the following operation:

db.getCollection("_foo").insert( { a : 1 } )

To perform a query, use the find() method, in as the following:

db.getCollection("_foo").find()

10.2.17 How do I isolate cursors from intervening write operations?

MongoDB cursors can return the same document more than once in some situations. 10 You can use the snapshot()
method on a cursor to isolate the operation for a very specific case.
snapshot() traverses the index on the _id field and guarantees that the query will return each document (with
respect to the value of the _id field) no more than once. 11
The snapshot() does not guarantee that the data returned by the query will reflect a single moment in time nor
does it provide isolation from insert or delete operations.
9 http://api.mongodb.org/
10

As a cursor returns documents other operations may interleave with the query: if some of these operations are updates (page 42) that cause the
document to move (in the case of a table scan, caused by document growth) or that change the indexed field on the index used by the query; then
the cursor will return the same document more than once.
11 MongoDB does not permit changes to the value of the _id field; it is not possible for a cursor that transverses this index to pass the same
document more than once.

10.2. FAQ: MongoDB for Application Developers

565

MongoDB Documentation, Release 2.4.9

Warning:
You cannot use snapshot() with sharded collections.
You cannot use snapshot() with sort() or hint() cursor methods.
As an alternative, if your collection has a field or fields that are never modified, you can use a unique index on this
field or these fields to achieve a similar result as the snapshot(). Query with hint() to explicitly force the query
to use that index.

10.2.18 When should I embed documents within other documents?

When modeling data in MongoDB (page 99), embedding is frequently the choice for:
contains relationships between entities.
one-to-many relationships when the many objects always appear with or are viewed in the context of their
parents.
You should also consider embedding for performance reasons if you have a collection with a large number of small
documents. Nevertheless, if small, separate documents represent the natural model for the data, then you should
maintain that model.
If, however, you can group these small documents by some logical relationship and you frequently retrieve the documents by this grouping, you might consider rolling-up the small documents into larger documents that contain an
array of subdocuments. Keep in mind that if you often only need to retrieve a subset of the documents within the
group, then rolling-up the documents may not provide better performance.
Rolling up these small documents into logical groupings means that queries to retrieve a group of documents involve
sequential reads and fewer random disk accesses.
Additionally, rolling up documents and moving common fields to the larger document benefit the index on these
fields. There would be fewer copies of the common fields and there would be fewer associated key entries in the
corresponding index. See Index Concepts (page 318) for more information on indexes.

10.2.19 Where can I learn more about data modeling in MongoDB?

Begin by reading the documents in the Data Models (page 97) section. These documents contain a high level introduction to data modeling considerations in addition to practical examples of data models targeted at particular issues.
Additionally, consider the following external resources that provide additional examples:
Schema Design by Example12
Dynamic Schema Blog Post13
MongoDB Data Modeling and Rails14
Ruby Example of Materialized Paths15
Sean Cribs Blog Post16 which was the source for much of the data-modeling-trees content.
12 http://www.10gen.com/presentations/mongodb-melbourne-2012/schema-design-example
13 http://dmerr.tumblr.com/post/6633338010/schemaless
14 http://docs.mongodb.org/ecosystem/tutorial/model-data-for-ruby-on-rails/
15 http://github.com/banker/newsmonger/blob/master/app/models/comment.rb
16 http://seancribbs.com/tech/2009/09/28/modeling-a-tree-in-a-document-database

566

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

10.2.20 Can I manually pad documents to prevent moves during updates?

An update can cause a document to move on disk if the document grows in size. To minimize document movements,
MongoDB uses padding (page 57).
You should not have to pad manually because MongoDB adds padding automatically (page 57) and can adaptively
adjust the amount of padding added to documents to prevent document relocations following updates. You can change
the default paddingFactor calculation by using the collMod command with the usePowerOf2Sizes flag.
The usePowerOf2Sizes flag ensures that MongoDB allocates document space in sizes that are powers of 2, which
helps ensure that MongoDB can efficiently reuse free space created by document deletion or relocation.
However, if you must pad a document manually, you can add a temporary field to the document and then $unset the
field, as in the following example.
Warning: Do not manually pad documents in a capped collection. Applying manual padding to a document in a
capped collection can break replication. Also, the padding is not preserved if you re-sync the MongoDB instance.
var myTempPadding = [ "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"];
db.myCollection.insert( { _id: 5, paddingField: myTempPadding } );
db.myCollection.update( { _id: 5 },
{ $unset: { paddingField: "" } }
)
db.myCollection.update( { _id: 5 },
{ $set: { realField: "Some text that I might have needed padding for" } }
)

See also:
Padding Factor (page 57)

10.3 FAQ: The mongo Shell

10.3.1 How can I enter multi-line operations in the mongo shell?
If you end a line with an open parenthesis ((), an open brace ({), or an open bracket ([), then the subsequent
lines start with ellipsis ("...") until you enter the corresponding closing parenthesis ()), the closing brace (})
or the closing bracket (]). The mongo shell waits for the closing parenthesis, closing brace, or the closing bracket
before evaluating the code, as in the following example:
> if ( x > 0 ) {
... count++;
... print (x);
... }

You can exit the line continuation mode if you enter two blank lines, as in the following example:
> if (x > 0
...
...
>

10.3. FAQ: The mongo Shell

567

MongoDB Documentation, Release 2.4.9

10.3.2 How can I access different databases temporarily?

You can use db.getSiblingDB() method to access another database without switching databases, as in the following example which first switches to the test database and then accesses the sampleDB database from the test
database:
use test
db.getSiblingDB('sampleDB').getCollectionNames();

10.3.3 Does the mongo shell support tab completion and other keyboard shortcuts?
The mongo shell supports keyboard shortcuts. For example,
Use the up/down arrow keys to scroll through command history. See .dbshell documentation for more information on the .dbshell file.
Use <Tab> to autocomplete or to list the completion possibilities, as in the following example which uses
<Tab> to complete the method name starting with the letter c:
db.myCollection.c<Tab>

Because there are many collection methods starting with the letter c, the <Tab> will list the various methods
that start with c.
For a full list of the shortcuts, see Shell Keyboard Shortcuts

10.3.4 How can I customize the mongo shell prompt?

New in version 1.9.
You can change the mongo shell prompt by setting the prompt variable. This makes it possible to display additional
information in the prompt.
Set prompt to any string or arbitrary JavaScript code that returns a string, consider the following examples:
Set the shell prompt to display the hostname and the database issued:
var host = db.serverStatus().host;
var prompt = function() { return db+"@"+host+"> "; }

The mongo shell prompt should now reflect the new prompt:
[email protected]>

Set the shell prompt to display the database statistics:

var prompt = function() {
return "Uptime:"+db.serverStatus().uptime+" Documents:"+db.stats().objects+" > ";
}

The mongo shell prompt should now reflect the new prompt:
Uptime:1052 Documents:25024787 >

You can add the logic for the prompt in the .mongorc.js file to set the prompt each time you start up the mongo shell.

568

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

10.3.5 Can I edit long shell operations with an external text editor?
You can use your own editor in the mongo shell by setting the EDITOR environment variable before starting the
mongo shell. Once in the mongo shell, you can edit with the specified editor by typing edit <variable> or
edit <function>, as in the following example:
1. Set the EDITOR variable from the command line prompt:
EDITOR=vim

2. Start the mongo shell:

mongo

3. Define a function myFunction:

function myFunction () { }

4. Edit the function using your editor:

edit myFunction

The command should open the vim edit session. Remember to save your changes.
5. Type myFunction to see the function definition:
myFunction

The result should be the changes from your saved edit:

function myFunction() {
print("This was edited");
}

10.4 FAQ: Concurrency

Changed in version 2.2.
MongoDB allows multiple clients to read and write a single corpus of data using a locking system to ensure that all
clients receive the same view of the data and to prevent multiple applications from modifying the exact same pieces
of data at the same time. Locks help guarantee that all writes to a single document occur either in full or not at all.
See also:
Presentation on Concurrency and Internals in 2.217

10.4.1 What type of locking does MongoDB use?

MongoDB uses a readers-writer 18 lock that allows concurrent reads access to a database but gives exclusive access to
a single write operation.
When a read lock exists, many read operations may use this lock. However, when a write lock exists, a single write
operation holds the lock exclusively, and no other read or write operations may share the lock.
Locks are writer greedy, which means writes have preference over reads. When both a read and write are waiting
for a lock, MongoDB grants the lock to the write.
17 http://www.mongodb.com/presentations/concurrency-internals-mongodb-2-2
18

You may be familiar with a readers-writer lock as multi-reader or shared exclusive lock. See the Wikipedia page on Readers-Writer
Locks (http://en.wikipedia.org/wiki/Readers%E2%80%93writer_lock) for more information.

10.4. FAQ: Concurrency

569

MongoDB Documentation, Release 2.4.9

10.4.2 How granular are locks in MongoDB?

Changed in version 2.2.
Beginning with version 2.2, MongoDB implements locks on a per-database basis for most read and write operations.
Some global operations, typically short lived operations involving multiple databases, still require a global instance
wide lock. Before 2.2, there is only one global lock per mongod instance.
For example, if you have six databases and one takes a write lock, the other five are still available for read and write.

10.4.3 How do I see the status of locks on my mongod instances?

For reporting on lock utilization information on locks, use any of the following methods:
db.serverStatus(),
db.currentOp(),
mongotop,
mongostat, and/or
the MongoDB Management Service (MMS)19
Specifically, the locks document in the output of serverStatus, or the locks field in the current
operation reporting provides insight into the type of locks and amount of lock contention in your mongod
instance.
To terminate an operation, use db.killOp().

10.4.4 Does a read or write operation ever yield the lock?

In some situations, read and write operations can yield their locks.
Long running read and write operations, such as queries, updates, and deletes, yield under many conditions. In
MongoDB 2.0, operations yielded based on time slices and the number of operations waiting for the actively held
lock. After 2.2, more adaptive algorithms allow operations to yield based on predicted disk access (i.e. page faults).
New in version 2.0: Read and write operations will yield their locks if the mongod receives a page fault or fetches
data that is unlikely to be in memory. Yielding allows other operations that only need to access documents that are
already in memory to complete while mongod loads documents into memory.
Additionally, write operations that affect multiple documents (i.e. update() with the multi parameter) will yield
periodically to allow read operations during these long write operations. Similarly, long running read locks will yield
periodically to ensure that write operations have the opportunity to complete.
Changed in version 2.2: The use of yielding expanded greatly in MongoDB 2.2. Including the yield for page fault.
MongoDB tracks the contents of memory and predicts whether data is available before performing a read. If MongoDB
predicts that the data is not in memory a read operation yields its lock while MongoDB loads the data to memory. Once
data is available in memory, the read will reacquire the lock to complete the operation.

10.4.5 Which operations lock the database?

Changed in version 2.2.
The following table lists common database operations and the types of locks they use.
19 http://mms.mongodb.com/

570

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

Operation
Issue a query
Get more data from
a cursor
Insert data
Remove data
Update data
Map-reduce
Create an index
db.eval()
eval
aggregate()

Lock Type
Read lock
Read lock
Write lock
Write lock
Write lock
Read lock and write lock, unless operations are specified as non-atomic. Portions of
map-reduce jobs can run concurrently.
Building an index in the foreground, which is the default, locks the database for extended
periods of time.
Write lock. db.eval() blocks all other JavaScript processes.
Write lock. If used with the nolock lock option, the eval option does not take a write
lock and cannot write data to the database.
Read lock

10.4.6 Which administrative commands lock the database?

Certain administrative commands can exclusively lock the database for extended periods of time. In some deployments, for large databases, you may consider taking the mongod instance offline so that clients are not affected. For
example, if a mongod is part of a replica set, take the mongod offline and let other members of the set service load
while maintenance is in progress.
The following administrative operations require an exclusive (i.e. write) lock on the database for extended periods:
db.collection.ensureIndex(), when issued without setting background to true,
reIndex,
compact,
db.repairDatabase(),
db.createCollection(), when creating a very large (i.e. many gigabytes) capped collection,
db.collection.validate(), and
db.copyDatabase(). This operation may lock all databases. See Does a MongoDB operation ever lock
more than one database? (page 571).
The following administrative commands lock the database but only hold the lock for a very short time:
db.collection.dropIndex(),
db.getLastError(),
db.isMaster(),
rs.status() (i.e. replSetGetStatus),
db.serverStatus(),
db.auth(), and
db.addUser().

10.4.7 Does a MongoDB operation ever lock more than one database?
The following MongoDB operations lock multiple databases:
db.copyDatabase() must lock the entire mongod instance at once.
10.4. FAQ: Concurrency

571

MongoDB Documentation, Release 2.4.9

Journaling, which is an internal operation, locks all databases for short intervals. All databases share a single
journal.
User authentication (page 239) locks the admin database as well as the database the user is accessing.
All writes to a replica sets primary lock both the database receiving the writes and then the local database for
a short time. The lock for the local database allows the mongod to write to the primarys oplog and accounts
for a small portion of the total time of the operation.

10.4.8 How does sharding affect concurrency?

Sharding improves concurrency by distributing collections over multiple mongod instances, allowing shard servers
(i.e. mongos processes) to perform any number of operations concurrently to the various downstream mongod
instances.
Each mongod instance is independent of the others in the shard cluster and uses the MongoDB readers-writer lock
(page 569). The operations on one mongod instance do not block the operations on any others.

10.4.9 How does concurrency affect a replica set primary?

In replication, when MongoDB writes to a collection on the primary, MongoDB also writes to the primarys oplog,
which is a special collection in the local database. Therefore, MongoDB must lock both the collections database
and the local database. The mongod must lock both databases at the same time to keep the database valid and
ensure that write operations, even with replication, are all-or-nothing operations.

10.4.10 How does concurrency affect secondaries?

In replication, MongoDB does not apply writes serially to secondaries. Secondaries collect oplog entries in batches
and then apply those batches in parallel. Secondaries do not allow reads while applying the write operations, and apply
write operations in the order that they appear in the oplog.
MongoDB can apply several writes in parallel on replica set secondaries, in two phases:
1. During the first prefer phase, under a read lock, the mongod ensures that all documents affected by the operations are in memory. During this phase, other clients may execute queries against this member.
2. A thread pool using write locks applies all write operations in the batch as part of a coordinated write phase.

10.4.11 What kind of concurrency does MongoDB provide for JavaScript operations?
Changed in version 2.4: The V8 JavaScript engine added in 2.4 allows multiple JavaScript operations to run at the
same time. Prior to 2.4, a single mongod could only run a single JavaScript operation at once.

10.5 FAQ: Sharding with MongoDB

This document answers common questions about horizontal scaling using MongoDBs sharding.
If you dont find the answer youre looking for, check the complete list of FAQs (page 555) or post your question to
the MongoDB User Mailing List20 .
20 https://groups.google.com/forum/?fromgroups#!forum/mongodb-user

572

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

10.5.1 Is sharding appropriate for a new deployment?

Sometimes.
If your data set fits on a single server, you should begin with an unsharded deployment.
Converting an unsharded database to a sharded cluster is easy and seamless, so there is little advantage in configuring
sharding while your data set is small.
Still, all production deployments should use replica sets to provide high availability and disaster recovery.

10.5.2 How does sharding work with replication?

To use replication with sharding, deploy each shard as a replica set.

10.5.3 Can I change the shard key after sharding a collection?

No.
There is no automatic support in MongoDB for changing a shard key after sharding a collection. This reality underscores the importance of choosing a good shard key (page 492). If you must change a shard key after sharding a
collection, the best option is to:
dump all data from MongoDB into an external format.
drop the original sharded collection.
configure sharding using a more ideal shard key.
pre-split (page 536) the shard key range to ensure initial even distribution.
restore the dumped data into MongoDB.
See shardCollection, sh.shardCollection(), the Shard Key (page 492), Deploy a Sharded Cluster
(page 507), and SERVER-400021 for more information.

10.5.4 What happens to unsharded collections in sharded databases?

In the current implementation, all databases in a sharded cluster have a primary shard. All unsharded collection
within that database will reside on the same shard.

10.5.5 How does MongoDB distribute data across shards?

Sharding must be specifically enabled on a collection. After enabling sharding on the collection, MongoDB will assign
various ranges of collection data to the different shards in the cluster. The cluster automatically corrects imbalances
between shards by migrating ranges of data from one shard to another.

10.5.6 What happens if a client updates a document in a chunk during a migration?

The mongos routes the operation to the old shard, where it will succeed immediately. Then the shard mongod instances will replicate the modification to the new shard before the sharded cluster updates that chunks ownership,
which effectively finalizes the migration process.
21 https://jira.mongodb.org/browse/SERVER-4000

10.5. FAQ: Sharding with MongoDB

573

MongoDB Documentation, Release 2.4.9

10.5.7 What happens to queries if a shard is inaccessible or slow?

If a shard is inaccessible or unavailable, queries will return with an error.
However, a client may set the partial query bit, which will then return results from all available shards, regardless
of whether a given shard is unavailable.
If a shard is responding slowly, mongos will merely wait for the shard to return results.

10.5.8 How does MongoDB distribute queries among shards?

Changed in version 2.0.
The exact method for distributing queries to shards in a cluster depends on the nature of the query and the configuration
of the sharded cluster. Consider a sharded collection, using the shard key user_id, that has last_login and
email attributes:
For a query that selects one or more values for the user_id key:
mongos determines which shard or shards contains the relevant data, based on the cluster metadata, and directs
a query to the required shard or shards, and returns those results to the client.
For a query that selects user_id and also performs a sort:
mongos can make a straightforward translation of this operation into a number of queries against the relevant
shards, ordered by user_id. When the sorted queries return from all shards, the mongos merges the sorted
results and returns the complete result to the client.
For queries that select on last_login:
These queries must run on all shards: mongos must parallelize the query over the shards and perform a mergesort on the email of the documents found.

10.5.9 How does MongoDB sort queries in sharded environments?

If you call the cursor.sort() method on a query in a sharded environment, the mongod for each shard will sort
its results, and the mongos merges each shards results before returning them to the client.

10.5.10 How does MongoDB ensure unique _id field values when using a shard
key other than _id?
If you do not use _id as the shard key, then your application/client layer must be responsible for keeping the _id
field unique. It is problematic for collections to have duplicate _id values.
If youre not sharding your collection by the _id field, then you should be sure to store a globally unique identifier in
that field. The default BSON ObjectId (page 129) works well in this case.

10.5.11 Ive enabled sharding and added a second shard, but all the data is still on
one server. Why?
First, ensure that youve declared a shard key for your collection. Until you have configured the shard key, MongoDB
will not create chunks, and sharding will not occur.
Next, keep in mind that the default chunk size is 64 MB. As a result, in most situations, the collection needs to have at
least 64 MB of data before a migration will occur.

574

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

Additionally, the system which balances chunks among the servers attempts to avoid superfluous migrations. Depending on the number of shards, your shard key, and the amount of data, systems often require at least 10 chunks of data
to trigger migrations.
You can run db.printShardingStatus() to see all the chunks present in your cluster.

10.5.12 Is it safe to remove old files in the moveChunk directory?

Yes. mongod creates these files as backups during normal shard balancing operations.
Once these migrations are complete, you may delete these files.

10.5.13 How does mongos use connections?

Each client maintains a connection to a mongos instance. Each mongos instance maintains a pool of connections to
the members of a replica set supporting the sharded cluster. Clients use connections between mongos and mongod
instances one at a time. Requests are not multiplexed or pipelined. When client requests complete, the mongos
returns the connection to the pool.
See the System Resource Utilization (page 224) section of the UNIX ulimit Settings (page 223) document.

10.5.14 Why does mongos hold connections open?

mongos uses a set of connection pools to communicate with each shard. These pools do not shrink when the number
of clients decreases.
This can lead to an unused mongos with a large number of open connections. If the mongos is no longer in use, it is
safe to restart the process to close existing connections.

10.5.15 Where does MongoDB report on connections used by mongos?

Connect to the mongos with the mongo shell, and run the following command:
db._adminCommand("connPoolStats");

10.5.16 What does writebacklisten in the log mean?

The writeback listener is a process that opens a long poll to relay writes back from a mongod or mongos after
migrations to make sure they have not gone to the wrong server. The writeback listener sends writes back to the
correct server if necessary.
These messages are a key part of the sharding infrastructure and should not cause concern.

10.5.17 How should administrators deal with failed migrations?

Failed migrations require no administrative intervention. Chunk migrations always preserve a consistent state. If a migration fails to complete for some reason, the cluster retries the operation. When the migration completes successfully,
the data resides only on the new shard.

10.5. FAQ: Sharding with MongoDB

575

MongoDB Documentation, Release 2.4.9

10.5.18 What is the process for moving, renaming, or changing the number of config servers?
See Sharded Cluster Tutorials (page 506) for information on migrating and replacing config servers.

10.5.19 When do the mongos servers detect config server changes?

mongos instances maintain a cache of the config database that holds the metadata for the sharded cluster. This
metadata includes the mapping of chunks to shards.
mongos updates its cache lazily by issuing a request to a shard and discovering that its metadata is out of date. There
is no way to control this behavior from the client, but you can run the flushRouterConfig command against any
mongos to force it to refresh its cache.

10.5.20 Is it possible to quickly update mongos servers after updating a replica set
configuration?
The mongos instances will detect these changes without intervention over time. However, if you want to force the
mongos to reload its configuration, run the flushRouterConfig command against to each mongos directly.

10.5.21 What does the maxConns setting on mongos do?

The maxConns option limits the number of connections accepted by mongos.
If your client driver or application creates a large number of connections but allows them to time out rather than closing
them explicitly, then it might make sense to limit the number of connections at the mongos layer.
Set maxConns to a value slightly higher than the maximum number of connections that the client creates, or the
maximum size of the connection pool. This setting prevents the mongos from causing connection spikes on the
individual shards. Spikes like these may disrupt the operation and memory allocation of the sharded cluster.

10.5.22 How do indexes impact queries in sharded systems?

If the query does not include the shard key, the mongos must send the query to all shards as a scatter/gather
operation. Each shard will, in turn, use either the shard key index or another more efficient index to fulfill the query.
If the query includes multiple sub-expressions that reference the fields indexed by the shard key and the secondary
index, the mongos can route the queries to a specific shard and the shard will use the index that will allow it to fulfill
most efficiently. See this presentation22 for more information.

10.5.23 Can shard keys be randomly generated?

Shard keys can be random. Random keys ensure optimal distribution of data across the cluster.
Sharded clusters, attempt to route queries to specific shards when queries include the shard key as a parameter, because
these directed queries are more efficient. In many cases, random keys can make it difficult to direct queries to specific
shards.
22 http://www.slideshare.net/mongodb/how-queries-work-with-sharding

576

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

10.5.24 Can shard keys have a non-uniform distribution of values?

Yes. There is no requirement that documents be evenly distributed by the shard key.
However, documents that have the same shard key must reside in the same chunk and therefore on the same server. If
your sharded data set has too many documents with the exact same shard key you will not be able to distribute those
documents across your sharded cluster.

10.5.25 Can you shard on the _id field?

You can use any field for the shard key. The _id field is a common shard key.
Be aware that ObjectId() values, which are the default value of the _id field, increment as a timestamp. As a
result, when used as a shard key, all new documents inserted into the collection will initially belong to the same chunk
on a single shard. Although the system will eventually divide this chunk and migrate its contents to distribute data
more evenly, at any moment the cluster can only direct insert operations at a single shard. This can limit the throughput
of inserts. If most of your write operations are updates, this limitation should not impact your performance. However,
if you have a high insert volume, this may be a limitation.
To address this issue, MongoDB 2.4 provides hashed shard keys (page 493).

10.5.26 What do moveChunk commit failed errors mean?

At the end of a chunk migration (page 503), the shard must connect to the config database to update the chunks record
in the cluster metadata. If the shard fails to connect to the config database, MongoDB reports the following error:
ERROR: moveChunk commit failed: version is at <n>|<nn> instead of
<N>|<NN>" and "ERROR: TERMINATING"

When this happens, the primary member of the shards replica set then terminates to protect data consistency. If a
secondary member can access the config database, data on the shard becomes accessible again after an election.
The user will need to resolve the chunk migration failure independently. If you encounter this issue, contact the
MongoDB User Group23 or MongoDB Support24 to address this issue.

10.5.27 How does draining a shard affect the balancing of uneven chunk distribution?
The sharded cluster balancing process controls both migrating chunks from decommissioned shards (i.e. draining) and
normal cluster balancing activities. Consider the following behaviors for different versions of MongoDB in situations
where you remove a shard in a cluster with an uneven chunk distribution:
After MongoDB 2.2, the balancer first removes the chunks from the draining shard and then balances the remaining uneven chunk distribution.
Before MongoDB 2.2, the balancer handles the uneven chunk distribution and then removes the chunks from
the draining shard.
23 http://groups.google.com/group/mongodb-user
24 http://www.mongodb.org/about/support

10.5. FAQ: Sharding with MongoDB

577

MongoDB Documentation, Release 2.4.9

10.6 FAQ: Replica Sets and Replication in MongoDB

This document answers common questions about database replication in MongoDB.
If you dont find the answer youre looking for, check the complete list of FAQs (page 555) or post your question to
the MongoDB User Mailing List25 .

10.6.1 What kinds of replication does MongoDB support?

MongoDB supports master-slave replication and a variation on master-slave replication known as replica sets. Replica
sets are the recommended replication topology.

10.6.2 What do the terms primary and master mean?

Primary and master nodes are the nodes that can accept writes. MongoDBs replication is single-master: only one
node can accept write operations at a time.
In a replica set, if the current primary node fails or becomes inaccessible, the other members can autonomously elect
one of the other members of the set to be the new primary.
By default, clients send all reads to the primary; however, read preference is configurable at the client level on a
per-connection basis, which makes it possible to send reads to secondary nodes instead.

10.6.3 What do the terms secondary and slave mean?

Secondary and slave nodes are read-only nodes that replicate from the primary.
Replication operates by way of an oplog, from which secondary/slave members apply new operations to themselves.
This replication process is asynchronous, so secondary/slave nodes may not always reflect the latest writes to the
primary. But usually, the gap between the primary and secondary nodes is just few milliseconds on a local network
connection.

10.6.4 How long does replica set failover take?

It varies, but a replica set will select a new primary within a minute.
It may take 10-30 seconds for the members of a replica set to declare a primary inaccessible. This triggers an election.
During the election, the cluster is unavailable for writes.
The election itself may take another 10-30 seconds.
Note: Eventually consistent reads, like the ones that will return from a replica set are only possible with a write
concern that permits reads from secondary members.

10.6.5 Does replication work over the Internet and WAN connections?
Yes.
For example, a deployment may maintain a primary and secondary in an East-coast data center along with a secondary
member for disaster recovery in a West-coast data center.
25 https://groups.google.com/forum/?fromgroups#!forum/mongodb-user

578

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

See also:
Deploy a Geographically Redundant Replica Set (page 425)

10.6.6 Can MongoDB replicate over a noisy connection?

Yes, but not without connection failures and the obvious latency.
Members of the set will attempt to reconnect to the other members of the set in response to networking flaps. This
does not require administrator intervention. However, if the network connections among the nodes in the replica set
are very slow, it might not be possible for the members of the node to keep up with the replication.
If the TCP connection between the secondaries and the primary instance breaks, a replica set will automatically elect
one of the secondary members of the set as primary.

10.6.7 What is the preferred replication method: master/slave or replica sets?

New in version 1.8.
Replica sets are the preferred replication mechanism in MongoDB. However, if your deployment requires more than
12 nodes, you must use master/slave replication.

10.6.8 What is the preferred replication method: replica sets or replica pairs?
Deprecated since version 1.6.
Replica sets replaced replica pairs in version 1.6. Replica sets are the preferred replication mechanism in MongoDB.

10.6.9 Why use journaling if replication already provides data redundancy?

Journaling facilitates faster crash recovery. Prior to journaling, crashes often required database repairs or full
data resync. Both were slow, and the first was unreliable.
Journaling is particularly useful for protection against power failures, especially if your replica set resides in a single
data center or power circuit.
When a replica set runs with journaling, mongod instances can safely restart without any administrator intervention.
Note: Journaling requires some resource overhead for write operations. Journaling has no effect on read performance,
however.
Journaling is enabled by default on all 64-bit builds of MongoDB v2.0 and greater.

10.6.10 Are write operations durable if write concern does not acknowledge
writes?
Yes.
However, if you want confirmation that a given write has arrived at the server, use write concern (page 47). The
getLastError command provides the facility for write concern. However, after the default write concern change
(page 634), the default write concern acknowledges all write operations, and unacknowledged writes must be explicitly configured. See the MongoDB Drivers and Client Libraries (page 95) documentation for your driver for more
information.

10.6. FAQ: Replica Sets and Replication in MongoDB

579

MongoDB Documentation, Release 2.4.9

10.6.11 How many arbiters do replica sets need?

Some configurations do not require any arbiter instances. Arbiters vote in elections for primary but do not replicate
the data like secondary members.
Replica sets require a majority of the remaining nodes present to elect a primary. Arbiters allow you to construct this
majority without the overhead of adding replicating nodes to the system.
There are many possible replica set architectures (page 390).
A replica set with an odd number of voting nodes does not need an arbiter.
A common configuration consists of two replicating nodes that include a primary and a secondary, as well as an
arbiter for the third node. This configuration makes it possible for the set to elect a primary in the event of failure,
without requiring three replicating nodes.
You may also consider adding an arbiter to a set if it has an equal number of nodes in two facilities and network
partitions between the facilities are possible. In these cases, the arbiter will break the tie between the two facilities and
allow the set to elect a new primary.
See also:
Replica Set Deployment Architectures (page 390)

10.6.12 What information do arbiters exchange with the rest of the replica set?
Arbiters never receive the contents of a collection but do exchange the following data with the rest of the replica set:
Credentials used to authenticate the arbiter with the replica set. All MongoDB processes within a replica set use
keyfiles. These exchanges are encrypted.
Replica set configuration data and voting data. This information is not encrypted. Only credential exchanges
are encrypted.
If your MongoDB deployment uses SSL, then all communications between arbiters and the other members of the
replica set are secure. See the documentation for Connect to MongoDB with SSL (page 252) for more information.
Run all arbiters on secure networks, as with all MongoDB components.
See
The overview of Arbiter Members of Replica Sets (page ??).

10.6.13 Which members of a replica set vote in elections?

All members of a replica set, unless the value of votes (page 469) is equal to 0, vote in elections. This includes all
delayed (page 388), hidden (page 387) and secondary-only (page 386) members, as well as the arbiters (page ??).
Additionally, the state of the voting members also determine whether the member can vote. Only voting members
in the following states are eligible to vote:
PRIMARY
SECONDARY
RECOVERING
ARBITER
ROLLBACK

580

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

See also:
Replica Set Elections (page 397)

10.6.14 Do hidden members vote in replica set elections?

Hidden members (page 387) of replica sets do vote in elections. To exclude a member from voting in an election,
change the value of the members votes (page 469) configuration to 0.
See also:
Replica Set Elections (page 397)

10.6.15 Is it normal for replica set members to use different amounts of disk space?
Yes.
Factors including: different oplog sizes, different levels of storage fragmentation, and MongoDBs data file preallocation can lead to some variation in storage utilization between nodes. Storage use disparities will be most pronounced when you add members at different times.

10.7 FAQ: MongoDB Storage

This document addresses common questions regarding MongoDBs storage system.
If you dont find the answer youre looking for, check the complete list of FAQs (page 555) or post your question to
the MongoDB User Mailing List26 .

10.7.1 What are memory mapped files?

A memory-mapped file is a file with data that the operating system places in memory by way of the mmap() system
call. mmap() thus maps the file to a region of virtual memory. Memory-mapped files are the critical piece of the
storage engine in MongoDB. By using memory mapped files MongoDB can treat the contents of its data files as if they
were in memory. This provides MongoDB with an extremely fast and simple method for accessing and manipulating
data.

10.7.2 How do memory mapped files work?

Memory mapping assigns files to a block of virtual memory with a direct byte-for-byte correlation. Once mapped, the
relationship between file and memory allows MongoDB to interact with the data in the file as if it were memory.

10.7.3 How does MongoDB work with memory mapped files?

MongoDB uses memory mapped files for managing and interacting with all data. MongoDB memory maps data files
to memory as it accesses documents. Data that isnt accessed is not mapped to memory.
26 https://groups.google.com/forum/?fromgroups#!forum/mongodb-user

10.7. FAQ: MongoDB Storage

581

MongoDB Documentation, Release 2.4.9

10.7.4 What are page faults?

Page faults will occur if youre attempting to access part of a memory-mapped file that isnt in memory.
If there is free memory, then the operating system can find the page on disk and load it to memory directly. However,
if there is no free memory, the operating system must:
find a page in memory that is stale or no longer needed, and write the page to disk.
read the requested page from disk and load it into memory.
This process, particularly on an active system can take a long time, particularly in comparison to reading a page that
is already in memory.

10.7.5 What is the difference between soft and hard page faults?
Page faults occur when MongoDB needs access to data that isnt currently in active memory. A hard page fault
refers to situations when MongoDB must access a disk to access the data. A soft page fault, by contrast, merely
moves memory pages from one list to another, such as from an operating system file cache. In production, MongoDB
will rarely encounter soft page faults.

10.7.6 What tools can I use to investigate storage use in MongoDB?

The db.stats() method in the mongo shell, returns the current state of the active database. The dbStats
command document describes the fields in the db.stats() output.

10.7.7 What is the working set?

Working set represents the total body of data that the application uses in the course of normal operation. Often this is
a subset of the total data size, but the specific size of the working set depends on actual moment-to-moment use of the
database.
If you run a query that requires MongoDB to scan every document in a collection, the working set will expand to
include every document. Depending on physical memory size, this may cause documents in the working set to page
out, or to be removed from physical memory by the operating system. The next time MongoDB needs to access these
documents, MongoDB may incur a hard page fault.
If you run a query that requires MongoDB to scan every document in a collection, the working set includes every
active document in memory.
For best performance, the majority of your active set should fit in RAM.

10.7.8 Why are the files in my data directory larger than the data in my database?
The data files in your data directory, which is the /data/db directory in default configurations, might be larger than
the data set inserted into the database. Consider the following possible causes:
Preallocated data files.
In the data directory, MongoDB preallocates data files to a particular size, in part to prevent file system fragmentation. MongoDB names the first data file <databasename>.0, the next <databasename>.1, etc.
The first file mongod allocates is 64 megabytes, the next 128 megabytes, and so on, up to 2 gigabytes, at which
point all subsequent files are 2 gigabytes. The data files include files with allocated space but that hold no data.
mongod may allocate a 1 gigabyte data file that may be 90% empty. For most larger databases, unused allocated
space is small compared to the database.
582

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

On Unix-like systems, mongod preallocates an additional data file and initializes the disk space to 0. Preallocating data files in the background prevents significant delays when a new database file is next allocated.
You can disable preallocation with the noprealloc run time option. However noprealloc is not intended
for use in production environments: only use noprealloc for testing and with small data sets where you
frequently drop databases.
On Linux systems you can use hdparm to get an idea of how costly allocation might be:
time hdparm --fallocate $((1024*1024)) testfile

The oplog.
If this mongod is a member of a replica set, the data directory includes the oplog.rs file, which is a preallocated
capped collection in the local database. The default allocation is approximately 5% of disk space on 64-bit
installations, see Oplog Sizing (page 411) for more information. In most cases, you should not need to resize
the oplog. However, if you do, see Change the Size of the Oplog (page 445).
The journal.
The data directory contains the journal files, which store write operations on disk prior to MongoDB applying
them to databases. See Journaling Mechanics (page 234).
Empty records.
MongoDB maintains lists of empty records in data files when deleting documents and collections. MongoDB
can reuse this space, but will never return this space to the operating system.
To de-fragment allocated storage, use compact, which de-fragments allocated space. By de-fragmenting storage, MongoDB can effectively use the allocated space. compact requires up to 2 gigabytes of extra disk space
to run. Do not use compact if you are critically low on disk space.
Important: compact only removes fragmentation from MongoDB data files and does not return any disk
space to the operating system.
To reclaim deleted space, use repairDatabase, which rebuilds the database which de-fragments the storage
and may release space to the operating system. repairDatabase requires up to 2 gigabytes of extra disk
space to run. Do not use repairDatabase if you are critically low on disk space.
Warning: repairDatabase requires enough free disk space to hold both the old and new database files
while the repair is running. Be aware that repairDatabase will block all other operations and may take
a long time to complete.

10.7.9 How can I check the size of a collection?

To view the size of a collection and other information, use the db.collection.stats() method from the mongo
shell. The following example issues db.collection.stats() for the orders collection:
db.orders.stats();

To view specific measures of size, use these methods:

db.collection.dataSize(): data size in bytes for the collection.
db.collection.storageSize(): allocation size in bytes, including unused space.
db.collection.totalSize(): the data size plus the index size in bytes.
db.collection.totalIndexSize(): the index size in bytes.

10.7. FAQ: MongoDB Storage

583

MongoDB Documentation, Release 2.4.9

Also, the following scripts print the statistics for each database and collection:

db._adminCommand("listDatabases").databases.forEach(function (d) {mdb = db.getSiblingDB(d.name); prin

db._adminCommand("listDatabases").databases.forEach(function (d) {mdb = db.getSiblingDB(d.name); mdb.

10.7.10 How can I check the size of indexes?

To view the size of the data allocated for an index, use one of the following procedures in the mongo shell:
Use the db.collection.stats() method using the index namespace. To retrieve a list of namespaces,
issue the following command:
db.system.namespaces.find()

Check the value of indexSizes in the output of the db.collection.stats() command.

Example
Issue the following command to retrieve index namespaces:
db.system.namespaces.find()

The command returns a list similar to the following:

{"name" : "test.orders"}
{"name" : "test.system.indexes"}
{"name" : "test.orders.$_id_"}

View the size of the data allocated for the orders.$_id_ index with the following sequence of operations:
use test
db.orders.$_id_.stats().indexSizes

10.7.11 How do I know when the server runs out of disk space?
If your server runs out of disk space for data files, you will see something like this in the log:
Thu
Thu
Thu
Thu
Thu
Thu

Aug
Aug
Aug
Aug
Aug
Aug

11
11
11
11
11
11

13:06:09
13:06:09
13:06:09
13:06:19
13:06:19
13:06:19

[FileAllocator]
[FileAllocator]
[FileAllocator]
[FileAllocator]
[FileAllocator]
[FileAllocator]

allocating new data file

error failed to allocate
will try again in 10
allocating new data file
error failed to allocate
will try again in 10

dbms/test.13, filling with zeroes...

new file: dbms/test.13 size: 2146435072
seconds
dbms/test.13, filling with zeroes...
new file: dbms/test.13 size: 2146435072
seconds

The server remains in this state forever, blocking all writes including deletes. However, reads still work. To delete
some data and compact, using the compact command, you must restart the server first.
If your server runs out of disk space for journal files, the server process will exit. By default, mongod creates journal
files in a sub-directory of dbpath named journal. You may elect to put the journal files on another storage device
using a filesystem mount or a symlink.
Note: If you place the journal files on a separate storage device you will not be able to use a file system snapshot tool
to capture a valid snapshot of your data files and journal files.

584

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

10.8 FAQ: Indexes

This document addresses common questions regarding MongoDB indexes.
If you dont find the answer youre looking for, check the complete list of FAQs (page 555) or post your question to
the MongoDB User Mailing List27 . See also Indexing Tutorials (page 339).

10.8.1 Should you run ensureIndex() after every insert?

No. You only need to create an index once for a single collection. After initial creation, MongoDB automatically
updates the index as data changes.
While running ensureIndex() is usually ok, if an index doesnt exist because of ongoing administrative work,
a call to ensureIndex() may disrupt database availability. Running ensureIndex() can render a replica set
inaccessible as the index creation is happening. See Build Indexes on Replica Sets (page 344).

10.8.2 How do you know what indexes exist in a collection?

To list a collections indexes, use the db.collection.getIndexes() method or a similar method for your
driver28 .

10.8.3 How do you determine the size of an index?

To check the sizes of the indexes on a collection, use db.collection.stats().

10.8.4 What happens if an index does not fit into RAM?

When an index is too large to fit into RAM, MongoDB must read the index from disk, which is a much slower operation
than reading from RAM. Keep in mind an index fits into RAM when your server has RAM available for the index
combined with the rest of the working set.
In certain cases, an index does not need to fit entirely into RAM. For details, see Indexes that Hold Only Recent Values
in RAM (page 373).

10.8.5 How do you know what index a query used?

To inspect how MongoDB processes a query, use the explain() method in the mongo shell, or in your application
driver.

10.8.6 How do you determine what fields to index?

A number of factors determine what fields to index, including selectivity (page 373), fitting indexes into RAM, reusing
indexes in multiple queries when possible, and creating indexes that can support all the fields in a given query. For
detailed documentation on choosing which fields to index, see Indexing Tutorials (page 339).
27 https://groups.google.com/forum/?fromgroups#!forum/mongodb-user
28 http://api.mongodb.org/

10.8. FAQ: Indexes

585

MongoDB Documentation, Release 2.4.9

10.8.7 How do write operations affect indexes?

Any write operation that alters an indexed field requires an update to the index in addition to the document itself. If
you update a document that causes the document to grow beyond the allotted record size, then MongoDB must update
all indexes that include this document as part of the update operation.
Therefore, if your application is write-heavy, creating too many indexes might affect performance.

10.8.8 Will building a large index affect database performance?

Building an index can be an IO-intensive operation, especially if you have a large collection. This is true on any
database system that supports secondary indexes, including MySQL. If you need to build an index on a large collection,
consider building the index in the background. See Index Creation (page 336).
If you build a large index without the background option, and if doing so causes the database to stop responding, do
one of the following:
Wait for the index to finish building.
Kill the current operation (see db.killOp()). The partial index will be deleted.

10.8.9 Can I use index keys to constrain query matches?

You can use the min() and max() methods to constrain the results of the cursor returned from find() by using
index keys.

10.8.10 Using $ne and $nin in a query is slow. Why?

The $ne and $nin operators are not selective. See Create Queries that Ensure Selectivity (page 373). If you need to
use these, it is often best to make sure that an additional, more selective criterion is part of the query.

10.8.11 Can I use a multi-key index to support a query for a whole array?
Not entirely. The index can partially support these queries because it can speed the selection of the first element of
the array; however, comparing all subsequent items in the array cannot use the index and must scan the documents
individually.

10.8.12 How can I effectively use indexes strategy for attribute lookups?
For simple attribute lookups that dont require sorted result sets or range queries, consider creating a field that contains
an array of documents where each document has a field (e.g. attrib ) that holds a specific type of attribute. You can
index this attrib field.
For example, the attrib field in the following document allows you to add an unlimited number of attributes types:
{ _id : ObjectId(...),
attrib : [
{ k: "color", v: "red" },
{ k: "shape": v: "rectangle" },
{ k: "color": v: "blue" },
{ k: "avail": v: true }
]
}

586

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

Both of the following queries could use the same { "attrib.k":

1, "attrib.v":

1 } index:

db.mycollection.find( { attrib: { $elemMatch : { k: "color", v: "blue" } } } )

db.mycollection.find( { attrib: { $elemMatch : { k: "avail", v: true } } } )

10.9 FAQ: MongoDB Diagnostics

This document provides answers to common diagnostic questions and issues.
If you dont find the answer youre looking for, check the complete list of FAQs (page 555) or post your question to
the MongoDB User Mailing List29 .

10.9.1 Where can I find information about a mongod process that stopped running
unexpectedly?
If mongod shuts down unexpectedly on a UNIX or UNIX-based platform, and if mongod fails to log a shutdown or
error message, then check your system logs for messages pertaining to MongoDB. For example, for logs located in
/var/log/messages, use the following commands:
sudo grep mongod /var/log/messages
sudo grep score /var/log/messages

10.9.2 Does TCP keepalive time affect sharded clusters and replica sets?
If you experience socket errors between members of a sharded cluster or replica set, that do not have other reasonable causes, check the TCP keep alive value, which Linux systems store as the tcp_keepalive_time value. A
common keep alive period is 7200 seconds (2 hours); however, different distributions and OS X may have different
settings. For MongoDB, you will have better experiences with shorter keepalive periods, on the order of 300 seconds
(five minutes).
On Linux systems you can use the following operation to check the value of tcp_keepalive_time:
cat /proc/sys/net/ipv4/tcp_keepalive_time

You can change the tcp_keepalive_time value with the following operation:
echo 300 > /proc/sys/net/ipv4/tcp_keepalive_time

The new tcp_keepalive_time value takes effect without requiring you to restart the mongod or mongos
servers. When you reboot or restart your system you will need to set the new tcp_keepalive_time value, or
see your operating systems documentation for setting the TCP keepalive value persistently.
For OS X systems, issue the following command to view the keep alive setting:
sysctl net.inet.tcp.keepinit

To set a shorter keep alive period use the following invocation:

sysctl -w net.inet.tcp.keepinit=300

If your replica set or sharded cluster experiences keepalive-related issues, you must alter the
tcp_keepalive_time value on all machines hosting MongoDB processes. This includes all machines
hosting mongos or mongod servers.
29 https://groups.google.com/forum/?fromgroups#!forum/mongodb-user

10.9. FAQ: MongoDB Diagnostics

587

MongoDB Documentation, Release 2.4.9

Windows users should consider the Windows Server Technet Article on KeepAliveTime configuration30 for more
information on setting keep alive for MongoDB deployments on Windows systems.

10.9.3 What tools are available for monitoring MongoDB?

The MongoDB Management Services <http://mms.mongodb.com> includes monitoring. MMS Monitoring is a free,
hosted services for monitoring MongoDB deployments. A full list of third-party tools is available as part of the
Monitoring for MongoDB (page 138) documentation. Also consider the MMS Documentation31 .

10.9.4 Memory Diagnostics

Do I need to configure swap space?
Always configure systems to have swap space. Without swap, your system may not be reliant in some situations with
extreme memory constraints, memory leaks, or multiple programs using the same memory. Think of the swap space
as something like a steam release valve that allows the system to release extra pressure without affecting the overall
functioning of the system.
Nevertheless, systems running MongoDB do not need swap for routine operation. Database files are memory-mapped
(page 581) and should constitute most of your MongoDB memory use. Therefore, it is unlikely that mongod will ever
use any swap space in normal operation. The operating system will release memory from the memory mapped files
without needing swap and MongoDB can write data to the data files without needing the swap system.
What is working set and how can I estimate its size?
The working set for a MongoDB database is the portion of your data that clients access most often. You can estimate size of the working set, using the workingSet document in the output of serverStatus. To return
serverStatus with the workingSet document, issue a command in the following form:
db.runCommand( { serverStatus: 1, workingSet: 1 } )

Must my working set size fit RAM?

Your working set should stay in memory to achieve good performance. Otherwise many random disk IOs will occur,
and unless you are using SSD, this can be quite slow.
One area to watch specifically in managing the size of your working set is index access patterns. If you are inserting
into indexes at random locations (as would happen with ids that are randomly generated by hashes), you will continually be updating the whole index. If instead you are able to create your ids in approximately ascending order (for
example, day concatenated with a random id), all the updates will occur at the right side of the b-tree and the working
set size for index pages will be much smaller.
It is fine if databases and thus virtual size are much larger than RAM.
How do I calculate how much RAM I need for my application?
The amount of RAM you need depends on several factors, including but not limited to:
The relationship between database storage (page 581) and working set.
30 http://technet.microsoft.com/en-us/library/dd349797.aspx#BKMK_2
31 http://mms.mongodb.com/help/

588

Chapter 10. Frequently Asked Questions

MongoDB Documentation, Release 2.4.9

The operating systems cache strategy for LRU (Least Recently Used)
The impact of journaling (page 234)
The number or rate of page faults and other MMS gauges to detect when you need more RAM
MongoDB defers to the operating system when loading data into memory from disk. It simply memory maps
(page 581) all its data files and relies on the operating system to cache data. The OS typically evicts the leastrecently-used data from RAM when it runs low on memory. For example if clients access indexes more frequently
than documents, then indexes will more likely stay in RAM, but it depends on your particular usage.
To calculate how much RAM you need, you must calculate your working set size, or the portion of your data that
clients use most often. This depends on your access patterns, what indexes you have, and the size of your documents.
If page faults are infrequent, your working set fits in RAM. If fault rates rise higher than that, you risk performance
degradation. This is less critical with SSD drives than with spinning disks.
How do I read memory statistics in the UNIX top command
Because mongod uses memory-mapped files (page 581), the memory statistics in top require interpretation in a
special way. On a large database, VSIZE (virtual bytes) tends to be the size of the entire database. If the mongod
doesnt have other processes running, RSIZE (resident bytes) is the total memory of the machine, as this counts file
system cache contents.
For Linux systems, use the vmstat command to help determine how the system uses memory. On OS X systems use
vm_stat.

10.9.5 Sharded Cluster Diagnostics

The two most important factors in maintaining a successful sharded cluster are:
choosing an appropriate shard key (page 492) and
sufficient capacity to support current and future operations (page 489).
You can prevent most issues encountered with sharding by ensuring that you choose the best possible shard key for
your deployment and ensure that you are always adding additional capacity to your cluster well before the current
resources become saturated. Continue reading for specific issues you may encounter in a production environment.
In a new sharded cluster, why does all data remains on one shard?
Your cluster must have sufficient data for sharding to make sense. Sharding works by migrating chunks between the
shards until each shard has roughly the same number of chunks.
The default chunk size is 64 megabytes. MongoDB will not begin migrations until the imbalance of chunks in the
cluster exceeds the migration threshold (page 502). While the default chunk size is configurable with the chunkSize
setting, these behaviors help prevent unnecessary chunk migrations, which can degrade the performance of your cluster
as a whole.
If you have just deployed a sharded cluster, make sure that you have enough data to make sharding effective. If you do
not have sufficient data to create more than eight 64 megabyte chunks, then all data will remain on one shard. Either
lower the chunk size (page 504) setting, or add more data to the cluster.
As a related problem, the system will split chunks only on inserts or updates, which means that if you configure
sharding and do not continue to issue insert and update operations, the database will not create any chunks. You can
either wait until your application inserts data or split chunks manually (page 537).

10.9. FAQ: MongoDB Diagnostics

589

MongoDB Documentation, Release 2.4.9

Finally, if your shard key has a low cardinality (page 512), MongoDB may not be able to create sufficient splits among
the data.
Why would one shard receive a disproportion amount of traffic in a sharded cluster?
In some situations, a single shard or a subset of the cluster will receive a disproportionate portion of the traffic and
workload. In almost all cases this is the result of a shard key that does not effectively allow write scaling (page 493).
Its also possible that you have hot chunks. In this case, you may be able to solve the problem by splitting and then
migrating parts of these chunks.
In the worst case, you may have to consider re-sharding your data and choosing a different shard key (page 511) to
correct this pattern.
What can prevent a sharded cluster from balancing?
If you have just deployed your sharded cluster, you may want to consider the troubleshooting suggestions for a new
cluster where data remains on a single shard (page 589).
If the cluster was initially balanced, but later developed an uneven distribution of data, consider the following possible
causes:
You have deleted or removed a significant amount of data from the cluster. If you have added additional data, it
may have a different distribution with regards to its shard key.
Your shard key has low cardinality (page 512) and MongoDB cannot split the chunks any further.
Your data set is growing faster than the balancer can distribute data around the cluster. This is uncommon and
typically is the result of:
a balancing window (page 532) that is too short, given the rate of data growth.
an uneven distribution of write operations (page 493) that requires more data migration. You may have to
choose a different shard key to resolve this issue.
poor network connectivity between shards, which may lead to chunk migrations that take too long to
complete. Investigate your network configuration and interconnections between shards.
Why do chunk migrations affect sharded cluster performance?
If migrations impact your cluster or applications performance, consider the following options, depending on the nature
of the impact:
1. If migrations only interrupt your clusters sporadically, you can limit the balancing window (page 532) to prevent
balancing activity during peak hours. Ensure that there is enough time remaining to keep the data from becoming
out of balance again.
2. If the balancer is always migrating chunks to the detriment of overall cluster performance:
You may want to attempt decreasing the chunk size (page 539) to limit the size of the migration.
Your cluster may be over capacity, and you may want to attempt to add one or two shards (page 514) to
the cluster to distribute load.
Its also possible that your shard key causes your application to direct all writes to a single shard. This kind of activity
pattern can require the balancer to migrate most data soon after writing it. Consider redeploying your cluster with a
shard key that provides better write scaling (page 493).

590

Chapter 10. Frequently Asked Questions

CHAPTER 11

Release Notes

Always install the latest, stable version of MongoDB. See MongoDB Version Numbers (page 634) for more information.
See the following release notes for an account of the changes in major versions. Release notes also include instructions
for upgrade.

11.1 Current Stable Release

(2.4-series)

11.1.1 Release Notes for MongoDB 2.4

March 19, 2013
MongoDB 2.4 includes enhanced geospatial support, switch to V8 JavaScript engine, security enhancements, and text
search (beta) and hashed index.
Minor Releases
2.4.9 January 10, 2014

Fix for instances where mongos incorrectly reports a successful write SERVER-121461 .
Make non-primary read preferences consistent with slaveOK versioning logic SERVER-119712 .
Allow new sharded cluster connections to read from secondaries when primary is down SERVER-72463 .
All 2.4.9 improvements4 .
2.4.8 November 1, 2013

Increase future compatibility for 2.6 authorization features SERVER-114785 .

1 https://jira.mongodb.org/browse/SERVER-12146
2 https://jira.mongodb.org/browse/SERVER-11971
3 https://jira.mongodb.org/browse/SERVER-7246
4 https://jira.mongodb.org/issues/?jql=fixVersion%20%3D%20%222.4.9%22%20AND%20project%20%3D%20SERVER
5 https://jira.mongodb.org/browse/SERVER-11478

591

MongoDB Documentation, Release 2.4.9

Fix dbhash cache issue for config servers SERVER-114216 .

All 2.4.8 improvements7 .
2.4.7 October 21, 2013

Fixed over-aggressive caching of V8 Isolates SERVER-105968 .

Removed extraneous initial count during mapReduce SERVER-99079 .
Cache results of dbhash command SERVER-1102110 .
Fixed memory leak in aggregation SERVER-1055411 .
All 2.4.7 improvements12 .
2.4.6 August 20, 2013

Fix for possible loss of documents during the chunk migration process if a document in the chunk is very large
SERVER-1047813 .
Fix for C++ client shutdown issues SERVER-889114 .
Improved replication robustness in presence of high network latency SERVER-1008515 .
Improved Solaris support SERVER-983216 , SERVER-978617 , and SERVER-708018 .
All 2.4.6 improvements19 .
2.4.5 July 3, 2013

Fix for CVE-2013-4650 Improperly grant user system privileges on databases other than local SERVER-998320 .
Fix for CVE-2013-3969 Remotely triggered segmentation fault in Javascript engine SERVER-987821 .
Fix to prevent identical background indexes from being built SERVER-985622 .
Config server performance improvements SERVER-986423 and SERVER-544224 .
Improved initial sync resilience to network failure SERVER-985325 .
6 https://jira.mongodb.org/browse/SERVER-11421
7 https://jira.mongodb.org/issues/?jql=fixVersion%20%3D%20%222.4.8%22%20AND%20project%20%3D%20SERVER
8 https://jira.mongodb.org/browse/SERVER-10596
9 https://jira.mongodb.org/browse/SERVER-9907
10 https://jira.mongodb.org/browse/SERVER-11021
11 https://jira.mongodb.org/browse/SERVER-10554
12 https://jira.mongodb.org/issues/?jql=fixVersion%20%3D%20%222.4.7%22%20AND%20project%20%3D%20SERVER
13 https://jira.mongodb.org/browse/SERVER-10478
14 https://jira.mongodb.org/browse/SERVER-8891
15 https://jira.mongodb.org/browse/SERVER-10085
16 https://jira.mongodb.org/browse/SERVER-9832
17 https://jira.mongodb.org/browse/SERVER-9786
18 https://jira.mongodb.org/browse/SERVER-7080
19 https://jira.mongodb.org/issues/?jql=fixVersion%20%3D%20%222.4.6%22%20AND%20project%20%3D%20SERVER
20 https://jira.mongodb.org/browse/SERVER-9983
21 https://jira.mongodb.org/browse/SERVER-9878
22 https://jira.mongodb.org/browse/SERVER-9856
23 https://jira.mongodb.org/browse/SERVER-9864
24 https://jira.mongodb.org/browse/SERVER-5442
25 https://jira.mongodb.org/browse/SERVER-9853

592

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

All 2.4.5 improvements26 .

2.4.4 June 4, 2013

Performance fix for Windows version SERVER-972127

Fix for config upgrade failure SERVER-966128 .
Migration to Cyrus SASL library for MongoDB Enterprise SERVER-881329 .
All 2.4.4 improvements30 .
2.4.3 April 23, 2013

Fix for mongo shell ignoring modified objects _id field SERVER-938531 .
Fix for race condition in log rotation SERVER-473932 .
Fix for copydb command with authorization in a sharded cluster SERVER-909333 .
All 2.4.3 improvements34 .
2.4.2 April 17, 2013

Several V8 memory leak and performance fixes SERVER-926735 and SERVER-923036 .

Fix for upgrading sharded clusters SERVER-912537 .
Fix for high volume connection crash SERVER-901438 .
All 2.4.2 improvements39
2.4.1 April 17, 2013

Fix for losing index changes during initial sync SERVER-908740

All 2.4.1 improvements41 .
Major New Features
The following changes in MongoDB affect both standard and Enterprise editions:
26 https://jira.mongodb.org/issues/?jql=fixVersion%20%3D%20%222.4.5%22%20AND%20project%20%3D%20SERVER
27 https://jira.mongodb.org/browse/SERVER-9721
28 https://jira.mongodb.org/browse/SERVER-9661
29 https://jira.mongodb.org/browse/SERVER-8813
30 https://jira.mongodb.org/issues/?jql=fixVersion%20%3D%20%222.4.4%22%20AND%20project%20%3D%20SERVER
31 https://jira.mongodb.org/browse/SERVER-9385
32 https://jira.mongodb.org/browse/SERVER-4739
33 https://jira.mongodb.org/browse/SERVER-9093
34 https://jira.mongodb.org/issues/?jql=fixVersion%20%3D%20%222.4.3%22%20AND%20project%20%3D%20SERVER
35 https://jira.mongodb.org/browse/SERVER-9267
36 https://jira.mongodb.org/browse/SERVER-9230
37 https://jira.mongodb.org/browse/SERVER-9125
38 https://jira.mongodb.org/browse/SERVER-9014
39 https://jira.mongodb.org/issues/?jql=fixVersion%20%3D%20%222.4.2%22%20AND%20project%20%3D%20SERVER
40 https://jira.mongodb.org/browse/SERVER-9087
41 https://jira.mongodb.org/issues/?jql=fixVersion%20%3D%20%222.4.1%22%20AND%20project%20%3D%20SERVER

11.1. Current Stable Release

593

MongoDB Documentation, Release 2.4.9

Text Search

Add support for text search of content in MongoDB databases as a beta feature. See Text Indexes (page 332) for more
information.
Geospatial Support Enhancements

Add new 2dsphere index (page 328). The new index supports GeoJSON42 objects Point, LineString, and
Polygon. See 2dsphere Indexes (page 328) and Geospatial Indexes and Queries (page 326).
Introduce operators $geometry, $geoWithin and $geoIntersects to work with the GeoJSON data.
Hashed Index

Add new hashed index (page 333) to index documents using hashes of field values. When used to index a shard key,
the hashed index ensures an evenly distributed shard key. See also Hashed Shard Keys (page 493).
Improvements to the Aggregation Framework

Improve support for geospatial queries. See the $geoWithin operator and the $geoNear pipeline stage.
Improve sort efficiency when the $sort stage immediately precedes a $limit in the pipeline.
Add new operators $millisecond and $concat and modify how $min operator processes null values.
Changes to Update Operators

Add new $setOnInsert operator for use with upsert .

Enhance functionality of the $push operator, supporting its use with the $each, the $sort, and the $slice
modifiers.
Additional Limitations for Map-Reduce and $where Operations

The mapReduce command, group command, and the $where operator expressions cannot access certain global
functions or properties, such as db, that are available in the mongo shell. See the individual command or operator for
details.
Improvements to serverStatus Command

Provide additional metrics and customization for the serverStatus command. See db.serverStatus() and
serverStatus for more information.
42 http://geojson.org/geojson-spec.html

594

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

Security Enhancements
Introduce a role-based access control system User Privilege Roles in MongoDB (page 267) using new system.users Privilege Documents (page 272).
Enforce uniqueness of the user in user privilege documents per database. Previous versions of MongoDB did
not enforce this requirement, and existing databases may have duplicates.
Support encrypted connections using SSL certificates signed by a Certificate Authority. See Connect to MongoDB with SSL (page 252).
For more information on security and risk management strategies, see MongoDB Security Practices and Procedures
(page 237).
Performance Improvements
V8 JavaScript Engine

JavaScript Changes in MongoDB 2.4 Consider the following impacts of V8 JavaScript Engine (page 595) in MongoDB 2.4:
Tip
Use the new interpreterVersion() method in the mongo shell and the javascriptEngine field in the
output of db.serverBuildInfo() to determine which JavaScript engine a MongoDB binary uses.

Improved Concurrency Previously, MongoDB operations that required the JavaScript interpreter had to acquire
a lock, and a single mongod could only run a single JavaScript operation at a time. The switch to V8 improves
concurrency by permitting multiple JavaScript operations to run at the same time.
Modernized JavaScript Implementation (ES5) The 5th edition of ECMAscript43 , abbreviated as ES5, adds many
new language features, including:
standardized JSON44 ,
strict mode45 ,
function.bind()46 ,
array extensions47 , and
getters and setters.
With V8, MongoDB supports the ES5 implementation of Javascript with the following exceptions.
Note: The following features do not work as expected on documents returned from MongoDB queries:
Object.seal() throws an exception on documents returned from MongoDB queries.
Object.freeze() throws an exception on documents returned from MongoDB queries.
Object.preventExtensions() incorrectly allows the addition of new properties on documents returned
from MongoDB queries.
43 http://www.ecma-international.org/publications/standards/Ecma-262.htm
44 http://www.ecma-international.org/ecma-262/5.1/#sec-15.12.1
45 http://www.ecma-international.org/ecma-262/5.1/#sec-4.2.2
46 http://www.ecma-international.org/ecma-262/5.1/#sec-15.3.4.5
47 http://www.ecma-international.org/ecma-262/5.1/#sec-15.4.4.16

11.1. Current Stable Release

595

MongoDB Documentation, Release 2.4.9

enumerable properties, when added to documents returned from MongoDB queries, are not saved during
write operations.
See SERVER-821648 , SERVER-822349 , SERVER-821550 , and SERVER-821451 for more information.
For objects that have not been returned from MongoDB queries, the features work as expected.

Removed Non-Standard SpiderMonkey Features V8 does not support the following non-standard SpiderMonkey52 JavaScript extensions, previously supported by MongoDBs use of SpiderMonkey as its JavaScript engine.
E4X Extensions V8 does not support the non-standard E4X53 extensions. E4X provides a native XML54 object to
the JavaScript language and adds the syntax for embedding literal XML documents in JavaScript code.
You need to use alternative XML processing if you used any of the following constructors/methods:
XML()
Namespace()
QName()
XMLList()
isXMLName()
Destructuring Assignment V8 does not support the non-standard destructuring assignments. Destructuring assignment extract[s] data from arrays or objects using a syntax that mirrors the construction of array and object literals. Mozilla docs55
Example
The following destructuring assignment is invalid with V8 and throws a SyntaxError:
original = [4, 8, 15];
var [b, ,c] = a; // <== destructuring assignment
print(b) // 4
print(c) // 15

Iterator(), StopIteration(), and Generators V8 does not support Iterator(), StopIteration(), and generators56 .
InternalError() V8 does not support InternalError(). Use Error() instead.
48 https://jira.mongodb.org/browse/SERVER-8216
49 https://jira.mongodb.org/browse/SERVER-8223
50 https://jira.mongodb.org/browse/SERVER-8215
51 https://jira.mongodb.org/browse/SERVER-8214
52 https://developer.mozilla.org/en-US/docs/SpiderMonkey
53 https://developer.mozilla.org/en-US/docs/E4X
54 https://developer.mozilla.org/en-US/docs/E4X/Processing_XML_with_E4X
55 https://developer.mozilla.org/en-US/docs/JavaScript/New_in_JavaScript/1.7#Destructuring_assignment_(Merge_into_own_page.2Fsection)
56 https://developer.mozilla.org/en-US/docs/JavaScript/Guide/Iterators_and_Generators

596

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

for each...in Construct V8 does not support the use of for each...in57 construct. Use for (var x in y)
construct instead.
Example
The following for each (var x in y) construct is invalid with V8:
var o = { name: 'MongoDB', version: 2.4 };
for each (var value in o) {
print(value);
}

Instead, in version 2.4, you can use the for (var x in y) construct:
var o = { name: 'MongoDB', version: 2.4 };
for (var prop in o) {
var value = o[prop];
print(value);
}

You can also use the array instance method forEach() with the ES5 method Object.keys():
Object.keys(o).forEach(function (key) {
var value = o[key];
print(value);
});

Array Comprehension V8 does not support Array comprehensions58 .

Use other methods such as the Array instance methods map(), filter(), or forEach().
Example
With V8, the following array comprehension is invalid:
var a = { w: 1, x: 2, y: 3, z: 4 }
var arr = [i * i for each (i in a) if (i > 2)]
printjson(arr)

Instead, you can implement using the Array instance method forEach() and the ES5 method Object.keys()
:
var a = { w: 1, x: 2, y: 3, z: 4 }
var arr = [];
Object.keys(a).forEach(function (key) {
var val = a[key];
if (val > 2) arr.push(val * val);
})
printjson(arr)

Note:

The new logic uses the Array instance method forEach() and not the generic method

57 https://developer.mozilla.org/en-US/docs/JavaScript/Reference/Statements/for_each...in
58 https://developer.mozilla.org/en-US/docs/JavaScript/Guide/Predefined_Core_Objects#Array_comprehensions

11.1. Current Stable Release

597

MongoDB Documentation, Release 2.4.9

Array.forEach(); V8 does not support Array generic methods. See Array Generic Methods (page 599) for
more information.

Multiple Catch Blocks V8 does not support multiple catch blocks and will throw a SyntaxError.
Example
The following multiple catch blocks is invalid with V8 and will throw "SyntaxError:
if":

Unexpected token

try {
something()
} catch (err if err instanceof SomeError) {
print('some error')
} catch (err) {
print('standard error')
}

Conditional Function Definition V8 will produce different outcomes than SpiderMonkey with conditional function
definitions59 .
Example
The following conditional function definition produces different outcomes in SpiderMonkey versus V8:
function test () {
if (false) {
function go () {};
}
print(typeof go)
}

With SpiderMonkey, the conditional function outputs undefined, whereas with V8, the conditional function outputs
function.
If your code defines functions this way, it is highly recommended that you refactor the code. The following example
refactors the conditional function definition to work in both SpiderMonkey and V8.
function test () {
var go;
if (false) {
go = function () {}
}
print(typeof go)
}

The refactored code outputs undefined in both SpiderMonkey and V8.

Note: ECMAscript prohibits conditional function definitions. To force V8 to throw an Error, enable strict mode60 .
function test () {
'use strict';
59 https://developer.mozilla.org/en-US/docs/JavaScript/Guide/Functions
60 http://www.nczonline.net/blog/2012/03/13/its-time-to-start-using-javascript-strict-mode/

598

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

if (false) {
function go () {}
}
}

The JavaScript code throws the following syntax error:

SyntaxError: In strict mode code, functions can only be declared at top level or immediately within a

String Generic Methods V8 does not support String generics61 . String generics are a set of methods on the String
class that mirror instance methods.
Example
The following use of the generic method String.toLowerCase() is invalid with V8:
var name = 'MongoDB';
var lower = String.toLowerCase(name);

With V8, use the String instance method toLowerCase() available through an instance of the String class
instead:
var name = 'MongoDB';
var lower = name.toLowerCase();
print(name + ' becomes ' + lower);

With V8, use the String instance methods instead of following generic methods:
String.charAt()
String.charCodeAt()
String.concat()
String.endsWith()
String.indexOf()
String.lastIndexOf()
String.localeCompare()
String.match()

String.quote()
String.replace()
String.search()
String.slice()
String.split()
String.startsWith()
String.substr()
String.substring()

String.toLocaleLowerCase()
String.toLocaleUpperCase()
String.toLowerCase()
String.toUpperCase()
String.trim()
String.trimLeft()
String.trimRight()

Array Generic Methods V8 does not support Array generic methods62 . Array generics are a set of methods on the
Array class that mirror instance methods.
Example
The following use of the generic method Array.every() is invalid with V8:
var arr = [4, 8, 15, 16, 23, 42];
function isEven (val) {
return 0 === val % 2;
}
61 https://developer.mozilla.org/en-US/docs/JavaScript/Reference/Global_Objects/String#String_generic_methods
62 https://developer.mozilla.org/en-US/docs/JavaScript/Reference/Global_Objects/Array#Array_generic_methods

11.1. Current Stable Release

599

MongoDB Documentation, Release 2.4.9

var allEven = Array.every(arr, isEven);

print(allEven);

With V8, use the Array instance method every() available through an instance of the Array class instead:
var allEven = arr.every(isEven);
print(allEven);

With V8, use the Array instance methods instead of the following generic methods:
Array.concat()
Array.every()
Array.filter()
Array.forEach()
Array.indexOf()
Array.join()

Array.lastIndexOf()
Array.map()
Array.pop()
Array.push()
Array.reverse()
Array.shift()

Array.slice()
Array.some()
Array.sort()
Array.splice()
Array.unshift()

Array Instance Method toSource() V8 does not support the Array instance method toSource()63 . Use the
Array instance method toString() instead.
uneval() V8 does not support the non-standard method uneval(). Use the standardized JSON.stringify()64
method instead.
Change default JavaScript engine from SpiderMonkey to V8. The change provides improved concurrency for
JavaScript operations, modernized JavaScript implementation, and the removal of non-standard SpiderMonkey features, and affects all JavaScript behavior including the commands mapReduce, group, and eval and the query
operator $where.
See JavaScript Changes in MongoDB 2.4 (page 595) for more information about all changes .
BSON Document Validation Enabled by Default for mongod and mongorestore

Enable basic BSON object validation for mongod and mongorestore when writing to MongoDB data files. See
objcheck for details.
Index Build Enhancements

Add support for multiple concurrent index builds in the background by a single mongod instance. See building
indexes in the background (page 336) for more information on background index builds.
Allow the db.killOp() method to terminate a foreground index build.
Improve index validation during index creation. See Compatibility and Index Type Changes in MongoDB 2.4
(page 608) for more information.
Set Parameters as Command Line Options

Provide --setParameter as a command line option for mongos and mongod. See mongod and mongos for
list of available options for setParameter.
63 https://developer.mozilla.org/en-US/docs/JavaScript/Reference/Global_Objects/Array/toSource
64 https://developer.mozilla.org/en-US/docs/JavaScript/Reference/Global_Objects/JSON/stringify

600

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

Increased Chunk Migration Write Concern

The default write concern for insert and delete operations that occur as part of a chunk migration in a sharded cluster
now ensures that at least one secondary acknowledges each insert and deletion operation. See Chunk Migration Write
Concern (page 504).
Improved Chunk Migration Queue Behavior

Increase performance for moving multiple chunks off an overloaded shard. The balancer no longer waits for the
current migrations delete phase to complete before starting the next chunk migration. See Chunk Migration Queuing
(page 503) for details.
Enterprise
The following changes are specific to MongoDB Enterprise Editions:
SASL Library Change

In 2.4.4, MongoDB Enterprise uses Cyrus SASL. Earlier 2.4 Enterprise versions use GNU SASL (libgsasl). To
upgrade to 2.4.4 MongoDB Enterprise or greater, you must install all package dependencies related to this change,
including the appropriate Cyrus SASL GSSAPI library. See Install MongoDB Enterprise (page 16) for details of the
dependencies.
New Modular Authentication System with Support for Kerberos

In 2.4, the MongoDB Enterprise now supports authentication via a Kerberos mechanism. See Deploy MongoDB with
Kerberos Authentication (page 261) for more information. For drivers that provide support for Kerberos authentication
to MongoDB, refer to Use MongoDB Drivers to Authenticate with Kerberos (page 264).
For more information on security and risk management strategies, see MongoDB Security Practices and Procedures
(page 237).
Additional Information
Platform Notes

For OS X, MongoDB 2.4 only supports OS X versions 10.6 (Snow Leopard) and later. There are no other platform
support changes in MongoDB 2.4. See the downloads page65 for more information on platform support.
Upgrade Process

Upgrade MongoDB to 2.4 In the general case, the upgrade from MongoDB 2.2 to 2.4 is a binary-compatible dropin upgrade: shut down the mongod instances and replace them with mongod instances running 2.4. However, before
you attempt any upgrade please familiarize yourself with the content of this document, particularly the procedure for
upgrading sharded clusters (page 602) and the considerations for reverting to 2.2 after running 2.4 (page 607).
65 http://www.mongodb.org/downloads/

11.1. Current Stable Release

601

MongoDB Documentation, Release 2.4.9

Upgrade Recommendations and Checklist When upgrading, consider the following:

For all deployments using authentication, upgrade the drivers (i.e. client libraries), before upgrading the
mongod instance or instances.
To upgrade to 2.4 sharded clusters must upgrade following the meta-data upgrade procedure (page 602).
If youre using 2.2.0 and running with auth enabled, you will need to upgrade first to 2.2.1 and then upgrade
to 2.4. See Rolling Upgrade Limitation for 2.2.0 Deployments Running with auth Enabled (page 606).
If you have system.users (page 272) documents (i.e. for auth) that you created before 2.4 you must
ensure that there are no duplicate values for the user field in the system.users (page 272) collection in
any database. If you do have documents with duplicate user fields, you must remove them before upgrading.
See Security Enhancements (page 595) for more information.
Upgrade Standalone mongod Instance to MongoDB 2.4
1. Download binaries of the latest release in the 2.4 series from the MongoDB Download Page66 . See Install
MongoDB (page 3) for more information.
2. Shutdown your mongod instance. Replace the existing binary with the 2.4 mongod binary and restart mongod.
Upgrade a Replica Set from MongoDB 2.2 to MongoDB 2.4 You can upgrade to 2.4 by performing a rolling upgrade of the set by upgrading the members individually while the other members are available to minimize downtime.
Use the following procedure:
1. Upgrade the secondary members of the set one at a time by shutting down the mongod and replacing the 2.2
binary with the 2.4 binary. After upgrading a mongod instance, wait for the member to recover to SECONDARY
state before upgrading the next instance. To check the members state, issue rs.status() in the mongo
shell.
2. Use the mongo shell method rs.stepDown() to step down the primary to allow the normal failover
(page 397) procedure. rs.stepDown() expedites the failover procedure and is preferable to shutting down
the primary directly.
Once the primary has stepped down and another member has assumed PRIMARY state, as observed in the output
of rs.status(), shut down the previous primary and replace mongod binary with the 2.4 binary and start
the new process.
Note: Replica set failover is not instant but will render the set unavailable to read or accept writes until the
failover process completes. Typically this takes 10 seconds or more. You may wish to plan the upgrade during
a predefined maintenance window.

Upgrade a Sharded Cluster from MongoDB 2.2 to MongoDB 2.4

Important: Only upgrade sharded clusters to 2.4 if all members of the cluster are currently running instances of 2.2.
The only supported upgrade path for sharded clusters running 2.0 is via 2.2.

Overview Upgrading a sharded cluster from MongoDB version 2.2 to 2.4 (or 2.3) requires that you run a 2.4
mongos with the --upgrade option, described in this procedure. The upgrade process does not require downtime.
The upgrade to MongoDB 2.4 adds epochs to the meta-data for all collections and chunks in the existing cluster.
MongoDB 2.2 processes are capable of handling epochs, even though 2.2 did not require them. This procedure applies
66 http://www.mongodb.org/downloads

602

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

only to upgrades from version 2.2. Earlier versions of MongoDB do not correctly handle epochs. See Cluster Metadata Upgrade (page 603) for more information.
After completing the meta-data upgrade you can fully upgrade the components of the cluster. With the balancer
disabled:
Upgrade all mongos instances in the cluster.
Upgrade all 3 mongod config server instances.
Upgrade the mongod instances for each shard, one at a time.
See Upgrade Sharded Cluster Components (page 606) for more information.
Cluster Meta-data Upgrade
Considerations Beware of the following properties of the cluster upgrade process:
Before you start the upgrade, ensure that the amount of free space on the filesystem for the config database
(page 547) is at least 4 to 5 times the amount of space currently used by the config database (page 547) data
files.
Additionally, ensure that all indexes in the config database (page 547) are {v:1} indexes. If a critical index is
a {v:0} index, chunk splits can fail due to known issues with the {v:0} format. {v:0} indexes are present
on clusters created with MongoDB 2.0 or earlier.
The duration of the metadata upgrade depends on the network latency between the node performing the upgrade
and the three config servers. Ensure low latency between the upgrade process and the config servers.
While the upgrade is in progress, you cannot make changes to the collection meta-data. For example, during the
upgrade, do not perform:
sh.enableSharding(),
sh.shardCollection(),
sh.addShard(),
db.createCollection(),
db.collection.drop(),
db.dropDatabase(),
any operation that creates a database, or
any other operation that modifies the cluster meta-data in any way. See Sharding Reference (page 546) for
a complete list of sharding commands. Note, however, that not all commands on the Sharding Reference
(page 546) page modifies the cluster meta-data.
Once you upgrade to 2.4 and complete the upgrade procedure do not use 2.0 mongod and mongos processes
in your cluster. 2.0 process may re-introduce old meta-data formats into cluster meta-data.
The upgraded config database will require more storage space than before, to make backup and working copies of the
config.chunks (page 549) and config.collections (page 550) collections. As always, if storage requirements increase, the mongod might need to pre-allocate additional data files. See What tools can I use to investigate
storage use in MongoDB? (page 582) for more information.

11.1. Current Stable Release

603

MongoDB Documentation, Release 2.4.9

Meta-data Upgrade Procedure Changes to the meta-data format for sharded clusters, stored in the config database
(page 547), require a special meta-data upgrade procedure when moving to 2.4.
Do not perform operations that modify meta-data while performing this procedure. See Upgrade a Sharded Cluster
from MongoDB 2.2 to MongoDB 2.4 (page 602) for examples of prohibited operations.
1. Before you start the upgrade, ensure that the amount of free space on the filesystem for the config database
(page 547) is at least 4 to 5 times the amount of space currently used by the config database (page 547) data
files.
Additionally, ensure that all indexes in the config database (page 547) are {v:1} indexes. If a critical index is
a {v:0} index, chunk splits can fail due to known issues with the {v:0} format. {v:0} indexes are present
on clusters created with MongoDB 2.0 or earlier.
The duration of the metadata upgrade depends on the network latency between the node performing the upgrade
and the three config servers. Ensure low latency between the upgrade process and the config servers.
To check the version of your indexes, use db.collection.getIndexes().
If any index on the config database is {v:0}, you should rebuild those indexes by connecting to the mongos
and either: rebuild all indexes using the db.collection.reIndex() method, or drop and rebuild specific
indexes using db.collection.dropIndex() and then db.collection.ensureIndex(). If you
need to upgrade the _id index to {v:1} use db.collection.reIndex().
You may have {v:0} indexes on other databases in the cluster.
2. Turn off the balancer (page 501) in the sharded cluster, as described in Disable the Balancer (page 533).
Optional
For additional security during the upgrade, you can make a backup of the config database using mongodump
or other backup tools.
3. Ensure there are no version 2.0 mongod or mongos processes still active in the sharded cluster. The automated
upgrade process checks for 2.0 processes, but network availability can prevent a definitive check. Wait 5 minutes
after stopping or upgrading version 2.0 mongos processes to confirm that none are still active.
4. Start a single 2.4 mongos process with configdb pointing to the sharded clusters config servers (page 488)
and with the --upgrade option. The upgrade process happens before the process becomes a daemon (i.e.
before --fork.)
You can upgrade an existing mongos instance to 2.4 or you can start a new mongos instance that can reach all
config servers if you need to avoid reconfiguring a production mongos.
Start the mongos with a command that resembles the following:
mongos --configdb <config servers> --upgrade

Without the --upgrade option 2.4 mongos processes will fail to start until the upgrade process is complete.
The upgrade will prevent any chunk moves or splits from occurring during the upgrade process. If there are
very many sharded collections or there are stale locks held by other failed processes, acquiring the locks for all
collections can take seconds or minutes. See the log for progress updates.
5. When the mongos process starts successfully, the upgrade is complete. If the mongos process fails to start,
check the log for more information.
If the mongos terminates or loses its connection to the config servers during the upgrade, you may always
safely retry the upgrade.

604

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

However, if the upgrade failed during the short critical section, the mongos will exit and report that the upgrade will require manual intervention. To continue the upgrade process, you must follow the Resync after an
Interruption of the Critical Section (page 605) procedure.
Optional
If the mongos logs show the upgrade waiting for the upgrade lock, a previous upgrade process may still be
active or may have ended abnormally. After 15 minutes of no remote activity mongos will force the upgrade
lock. If you can verify that there are no running upgrade processes, you may connect to a 2.2 mongos process
and force the lock manually:
mongo <mongos.example.net>
db.getMongo().getCollection("config.locks").findOne({ _id : "configUpgrade" })

If the process specified in the process field of this document is verifiably offline, run the following operation
to force the lock.

db.getMongo().getCollection("config.locks").update({ _id : "configUpgrade" }, { $set : { state :

It is always more safe to wait for the mongos to verify that the lock is inactive, if you have any doubts about
the activity of another upgrade operation. In addition to the configUpgrade, the mongos may need to wait
for specific collection locks. Do not force the specific collection locks.
6. Upgrade and restart other mongos processes in the sharded cluster, without the --upgrade option.
See Upgrade Sharded Cluster Components (page 606) for more information.
7. Re-enable the balancer (page 533). You can now perform operations that modify cluster meta-data.
Once you have upgraded, do not introduce version 2.0 MongoDB processes into the sharded cluster. This can reintroduce old meta-data formats into the config servers. The meta-data change made by this upgrade process will help
prevent errors caused by cross-version incompatibilities in future versions of MongoDB.
Resync after an Interruption of the Critical Section During the short critical section of the upgrade that applies
changes to the meta-data, it is unlikely but possible that a network interruption can prevent all three config servers
from verifying or modifying data. If this occurs, the config servers (page 488) must be re-synced, and there may be
problems starting new mongos processes. The sharded cluster will remain accessible, but avoid all cluster metadata changes until you resync the config servers. Operations that change meta-data include: adding shards, dropping
databases, and dropping collections.
Note: Only perform the following procedure if something (e.g. network, power, etc.) interrupts the upgrade process
during the short critical section of the upgrade. Remember, you may always safely attempt the meta data upgrade
procedure (page 604).
To resync the config servers:
1. Turn off the balancer (page 501) in the sharded cluster and stop all meta-data operations. If you are in the
middle of an upgrade process (Upgrade a Sharded Cluster from MongoDB 2.2 to MongoDB 2.4 (page 602)),
you have already disabled the balancer.
2. Shut down two of the three config servers, preferably the last two listed in the configdb string. For example, if your configdb string is configA:27019,configB:27019,configC:27019, shut down
configB and configC. Shutting down the last two config servers ensures that most mongos instances will
have uninterrupted access to cluster meta-data.
3. mongodump the data files of the active config server (configA).

11.1. Current Stable Release

605

MongoDB Documentation, Release 2.4.9

4. Move the data files of the deactivated config servers (configB and configC) to a backup location.
5. Create new, empty data directories.
6. Restart the disabled config servers with --dbpath pointing to the now-empty data directory and --port
pointing to an alternate port (e.g. 27020).
7. Use mongorestore to repopulate the data files on the disabled documents from the active config server
(configA) to the restarted config servers on the new port (configB:27020,configC:27020). These
config servers are now re-synced.
8. Restart the restored config servers on the old port, resetting the port back to the old settings (configB:27019
and configC:27019).
9. In some cases connection pooling may cause spurious failures, as the mongos disables old connections only
after attempted use. 2.4 fixes this problem, but to avoid this issue in version 2.2, you can restart all mongos
instances (one-by-one, to avoid downtime) and use the rs.stepDown() method before restarting each of the
shard replica set primaries.
10. The sharded cluster is now fully resynced; however before you attempt the upgrade process again, you must
manually reset the upgrade state using a version 2.2 mongos. Begin by connecting to the 2.2 mongos with the
mongo shell:
mongo <mongos.example.net>

Then, use the following operation to reset the upgrade process:

db.getMongo().getCollection("config.version").update({ _id : 1 }, { $unset : { upgradeState : 1

11. Finally retry the upgrade process, as in Upgrade a Sharded Cluster from MongoDB 2.2 to MongoDB 2.4
(page 602).
Upgrade Sharded Cluster Components After you have successfully completed the meta-data upgrade process
described in Meta-data Upgrade Procedure (page 604), and the 2.4 mongos instance starts, you can upgrade the
other processes in your MongoDB deployment.
While the balancer is still disabled, upgrade the components of your sharded cluster in the following order:
Upgrade all mongos instances in the cluster, in any order.
Upgrade all 3 mongod config server instances, upgrading the first system in the mongos --configdb argument last.
Upgrade each shard, one at a time, upgrading the mongod secondaries before running replSetStepDown
and upgrading the primary of each shard.
When this process is complete, you can now re-enable the balancer (page 533).
Rolling Upgrade Limitation for 2.2.0 Deployments Running with auth Enabled MongoDB cannot support
deployments that mix 2.2.0 and 2.4.0, or greater, components. MongoDB version 2.2.1 and later processes can exist in
mixed deployments with 2.4-series processes. Therefore you cannot perform a rolling upgrade from MongoDB 2.2.0
to MongoDB 2.4.0. To upgrade a cluster with 2.2.0 components, use one of the following procedures.
1. Perform a rolling upgrade of all 2.2.0 processes to the latest 2.2-series release (e.g. 2.2.3) so that there are no
processes in the deployment that predate 2.2.1. When there are no 2.2.0 processes in the deployment, perform a
rolling upgrade to 2.4.0.
2. Stop all processes in the cluster. Upgrade all processes to a 2.4-series release of MongoDB, and start all processes at the same time.

606

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

Upgrade from 2.3 to 2.4 If you used a mongod from the 2.3 or 2.4-rc (release candidate) series, you can safely
transition these databases to 2.4.0 or later; however, if you created 2dsphere or text indexes using a mongod
before v2.4-rc2, you will need to rebuild these indexes. For example:
db.records.dropIndex( { loc: "2dsphere" } )
db.records.dropIndex( "records_text" )
db.records.ensureIndex( { loc: "2dsphere" } )
db.records.ensureIndex( { records: "text" } )

Downgrade MongoDB from 2.4 to Previous Versions For some cases the on-disk format of data files used by 2.4
and 2.2 mongod is compatible, and you can upgrade and downgrade if needed. However, several new features in 2.4
are incompatible with previous versions:
2dsphere indexes are incompatible with 2.2 and earlier mongod instances.
text indexes are incompatible with 2.2 and earlier mongod instances.
using a hashed index as a shard key are incompatible with 2.2 and earlier mongos instances.
hashed indexes are incompatible with 2.0 and earlier mongod instances.
Important: Collections sharded using hashed shard keys, should not use 2.2 mongod instances, which cannot
correctly support cluster operations for these collections.
If you completed the meta-data upgrade for a sharded cluster (page 602), you can safely downgrade to 2.2 MongoDB
processes. Do not use 2.0 processes after completing the upgrade procedure.
Note: In sharded clusters, once you have completed the meta-data upgrade procedure (page 602), you cannot use 2.0
mongod or mongos instances in the same cluster.
If you complete the meta-data upgrade, you can safely downgrade components in any order. When upgrade again,
always upgrade mongos instances before mongod instances.
Do not create 2dsphere or text indexes in a cluster that has 2.2 components.

Considerations and Compatibility If you upgrade to MongoDB 2.4, and then need to run MongoDB 2.2 with the
same data files, consider the following limitations.
If you use a hashed index as the shard key index, which is only possible under 2.4 you will not be able to
query data in this sharded collection. Furthermore, a 2.2 mongos cannot properly route an insert operation
for a collections sharded using a hashed index for the shard key index: any data that you insert using a 2.2
mongos, will not arrive on the correct shard and will not be reachable by future queries.
If you never create an 2dsphere or text index, you can move between a 2.4 and 2.2 mongod for a given
data set; however, after you create the first 2dsphere or text index with a 2.4 mongod you will need to run
a 2.2 mongod with the --upgrade option and drop any 2dsphere or text index.
Upgrade and Downgrade Procedures
Basic Downgrade and Upgrade

Except as described below, moving between 2.2 and 2.4 is a drop-in replacement:

stop the existing mongod, using the --shutdown option as follows:

mongod --dbpath /var/mongod/data --shutdown

11.1. Current Stable Release

607

MongoDB Documentation, Release 2.4.9

Replace /var/mongod/data with your MongoDB dbpath.

start the new mongod processes with the same dbpath setting, for example:
mongod --dbpath /var/mongod/data

Replace /var/mongod/data with your MongoDB dbpath.

Downgrade to 2.2 After Creating a 2dsphere or text Index If you have created 2dsphere or text indexes while running a 2.4 mongod instance, you can downgrade at any time, by starting the 2.2 mongod with the
--upgrade option as follows:
mongod --dbpath /var/mongod/data/ --upgrade

Then, you will need to drop any existing 2dsphere or text indexes using db.collection.dropIndex(),
for example:
db.records.dropIndex( { loc: "2dsphere" } )
db.records.dropIndex( "records_text" )

Warning: --upgrade will run repairDatabase on any database where you have created a 2dsphere or
text index, which will rebuild all indexes.

Troubleshooting Upgrade/Downgrade Operations If you do not use --upgrade, when you attempt to start a
2.2 mongod and you have created a 2dsphere or text index, mongod will return the following message:

'need to upgrade database index_plugin_upgrade with pdfile version 4.6, new version: 4.5 Not upgradin

While running 2.4, to check the data file version of a MongoDB database, use the following operation in the shell:
db.getSiblingDB('<databasename>').stats().dataFileVersion

The major data file 67 version for both 2.2 and 2.4 is 4, the minor data file version for 2.2 is 5 and the minor data file
version for 2.4 is 6 after you create a 2dsphere or text index.
Compatibility and Index Type Changes in MongoDB 2.4 In 2.4 MongoDB includes two new features related to
indexes that users upgrading to version 2.4 must consider, particularly with regard to possible downgrade paths. For
more information on downgrades, see Downgrade MongoDB from 2.4 to Previous Versions (page 607).
New Index Types In 2.4 MongoDB adds two new index types: 2dsphere and text. These index types do not
exist in 2.2, and for each database, creating a 2dsphere or text index, will upgrade the data-file version and make
that database incompatible with 2.2.
If you intend to downgrade, you should always drop all 2dsphere and text indexes before moving to 2.2.
You can use the downgrade procedure (page 607) to downgrade these databases and run 2.2 if needed, however this
will run a full database repair (as with repairDatabase) for all affected databases.
Index Type Validation In MongoDB 2.2 and earlier you could specify invalid index types that did not exist. In
these situations, MongoDB would create an ascending (e.g. 1) index. Invalid indexes include index types specified by
strings that do not refer to an existing index type, and all numbers other than 1 and -1. 68
67

The data file version (i.e. pdfile version) is independent and unrelated to the release version of MongoDB.
In 2.4, indexes that specify a type of "1" or "-1" (the strings "1" and "-1") will continue to exist, despite a warning on start-up. However,
a secondary in a replica set cannot complete an initial sync from a primary that has a "1" or "-1" index. Avoid all indexes with invalid types.
68

608

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

In 2.4, creating any invalid index will result in an error. Furthermore, you cannot create a 2dsphere or text index
on a collection if its containing database has any invalid index types. 1
Example
If you attempt to add an invalid index in MongoDB 2.4, as in the following:
db.coll.ensureIndex( { field: "1" } )

MongoDB will return the following error document:

{
"err" : "Unknown index plugin '1' in index { field: \"1\" }"
"code": 16734,
"n": <number>,
"connectionId": <number>,
"ok": 1
}

See Upgrade MongoDB to 2.4 (page 601) for full upgrade instructions.
Other Resources

MongoDB Downloads69 .
All JIRA issues resolved in 2.470 .
All Backwards incompatible changes71 .
All Third Party License Notices72 .
See http://docs.mongodb.org/manualrelease-notes/2.4-changes for an overview of all changes
in 2.4.
See also:
See MongoDB 2.5 Series Development Release Notes for more information about the upcoming release of MongoDB.

11.2 Previous Stable Releases

11.2.1 Release Notes for MongoDB 2.2
Upgrading
MongoDB 2.2 is a production release series and succeeds the 2.0 production release series.
MongoDB 2.0 data files are compatible with 2.2-series binaries without any special migration process. However,
always perform the upgrade process for replica sets and sharded clusters using the procedures that follow.
69 http://mongodb.org/downloads

70 https://jira.mongodb.org/secure/IssueNavigator.jspa?reset=true&jqlQuery=project+%3D+SERVER+AND+fixVersion+in+%28%222.3.2%22,+%222.3.1%22,+%222
rc0%22,+%222.4.0-rc1%22,+%222.4.0-rc2%22,+%222.4.0-rc3%22%29
71 https://jira.mongodb.org/secure/IssueNavigator.jspa?reset=true&jqlQuery=project+%3D+SERVER+AND+fixVersion+in+%28%222.3.2%22%2C+%222.3.1%22%2
rc0%22%2C+%222.4.0-rc1%22%2C+%222.4.0-rc2%22%2C+%222.4.0-rc3%22%29+AND+%22Backward+Breaking%22+in+%28+Rarely+%2C+sometimes%2C+yes
72 https://github.com/mongodb/mongo/blob/v2.4/distsrc/THIRD-PARTY-NOTICES

11.2. Previous Stable Releases

609

MongoDB Documentation, Release 2.4.9

Synopsis

mongod, 2.2 is a drop-in replacement for 2.0 and 1.8.

Check your driver (page 95) documentation for information regarding required compatibility upgrades, and
always run the recent release of your driver.
Typically, only users running with authentication, will need to upgrade drivers before continuing with the upgrade to 2.2.
For all deployments using authentication, upgrade the drivers (i.e. client libraries), before upgrading the
mongod instance or instances.
For all upgrades of sharded clusters:
turn off the balancer during the upgrade process. See the Disable the Balancer (page 533) section for more
information.
upgrade all mongos instances before upgrading any mongod instances.
Other than the above restrictions, 2.2 processes can interoperate with 2.0 and 1.8 tools and processes. You can safely
upgrade the mongod and mongos components of a deployment one by one while the deployment is otherwise operational. Be sure to read the detailed upgrade procedures below before upgrading production systems.
Upgrading a Standalone mongod

1. Download binaries of the latest release in the 2.2 series from the MongoDB Download Page73 .
2. Shutdown your mongod instance. Replace the existing binary with the 2.2 mongod binary and restart MongoDB.
Upgrading a Replica Set

You can upgrade to 2.2 by performing a rolling upgrade of the set by upgrading the members individually while the
other members are available to minimize downtime. Use the following procedure:
1. Upgrade the secondary members of the set one at a time by shutting down the mongod and replacing the 2.0
binary with the 2.2 binary. After upgrading a mongod instance, wait for the member to recover to SECONDARY
state before upgrading the next instance. To check the members state, issue rs.status() in the mongo
shell.
2. Use the mongo shell method rs.stepDown() to step down the primary to allow the normal failover
(page 397) procedure. rs.stepDown() expedites the failover procedure and is preferable to shutting down
the primary directly.
Once the primary has stepped down and another member has assumed PRIMARY state, as observed in the output
of rs.status(), shut down the previous primary and replace mongod binary with the 2.2 binary and start
the new process.
Note: Replica set failover is not instant but will render the set unavailable to read or accept writes until the
failover process completes. Typically this takes 10 seconds or more. You may wish to plan the upgrade during
a predefined maintenance window.
73 http://downloads.mongodb.org/

610

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

Upgrading a Sharded Cluster

Use the following procedure to upgrade a sharded cluster:

Disable the balancer (page 533).
Upgrade all mongos instances first, in any order.
Upgrade all of the mongod config server instances using the stand alone (page 610) procedure. To keep the
cluster online, be sure that at all times at least one config server is up.
Upgrade each shards replica set, using the upgrade procedure for replica sets (page 610) detailed above.
re-enable the balancer.
Note: Balancing is not currently supported in mixed 2.0.x and 2.2.0 deployments. Thus you will want to reach a
consistent version for all shards within a reasonable period of time, e.g. same-day. See SERVER-690274 for more
information.

Changes
Major Features

Aggregation Framework The aggregation framework makes it possible to do aggregation operations without needing to use map-reduce. The aggregate command exposes the aggregation framework, and the aggregate()
helper in the mongo shell provides an interface to these operations. Consider the following resources for background
on the aggregation framework and its use:
Documentation: Aggregation Concepts (page 281)
Reference: Aggregation Reference (page 308)
Examples: Aggregation Examples (page 292)
TTL Collections TTL collections remove expired data from a collection, using a special index and a background
thread that deletes expired documents every minute. These collections are useful as an alternative to capped collections
in some cases, such as for data warehousing and caching cases, including: machine generated event data, logs, and
session information that needs to persist in a database for only a limited period of time.
For more information, see the Expire Data from Collections by Setting TTL (page 163) tutorial.
Concurrency Improvements MongoDB 2.2 increases the servers capacity for concurrent operations with the following improvements:
1. DB Level Locking75
2. Improved Yielding on Page Faults76
3. Improved Page Fault Detection on Windows77
To reflect these changes, MongoDB now provides changed and improved reporting for concurrency and use, see locks
and server-status-record-stats in server status and see db.currentOp(), mongotop, and mongostat.
74 https://jira.mongodb.org/browse/SERVER-6902
75 https://jira.mongodb.org/browse/SERVER-4328
76 https://jira.mongodb.org/browse/SERVER-3357
77 https://jira.mongodb.org/browse/SERVER-4538

11.2. Previous Stable Releases

611

MongoDB Documentation, Release 2.4.9

Improved Data Center Awareness with Tag Aware Sharding MongoDB 2.2 adds additional support for geographic distribution or other custom partitioning for sharded collections in clusters. By using this tag aware sharding, you can automatically ensure that data in a sharded database system is always on specific shards. For example,
with tag aware sharding, you can ensure that data is closest to the application servers that use that data most frequently.
Shard tagging controls data location, and is complementary but separate from replica set tagging, which controls
read preference (page 406) and write concern (page 47). For example, shard tagging can pin all USA data to one
or more logical shards, while replica set tagging can control which mongod instances (e.g. production or
reporting) the application uses to service requests.
See the documentation for the following helpers in the mongo shell that support tagged sharding configuration:
sh.addShardTag()
sh.addTagRange()
sh.removeShardTag()
Also, see Tag Aware Sharding (page 540) and Manage Shard Tags (page 541).
Fully Supported Read Preference Semantics All MongoDB clients and drivers now support full read preferences
(page 406), including consistent support for a full range of read preference modes (page 476) and tag sets (page 408).
This support extends to the mongos and applies identically to single replica sets and to the replica sets for each shard
in a sharded cluster.
Additional read preference support now exists in the mongo shell using the readPref() cursor method.
Compatibility Changes

Authentication Changes MongoDB 2.2 provides more reliable and robust support for authentication clients, including drivers and mongos instances.
If your cluster runs with authentication:
For all drivers, use the latest release of your driver and check its release notes.
In sharded environments, to ensure that your cluster remains available during the upgrade process you must use
the upgrade procedure for sharded clusters (page 611).
findAndModify Returns Null Value for Upserts that Perform Inserts In version 2.2, for upsert that perform
inserts with the new option set to false, findAndModify commands will now return the following output:
{ 'ok': 1.0, 'value': null }

In the mongo shell, upsert findAndModify operations that perform inserts (with new set to false.)only output a
null value.
In version 2.0 these operations would return an empty document, e.g. { }.
See: SERVER-622678 for more information.
mongodump 2.2 Output Incompatible with Pre-2.2 mongorestore If you use the mongodump tool from the
2.2 distribution to create a dump of a database, you must use a 2.2 (or later) version of mongorestore to restore
that dump.
See: SERVER-696179 for more information.
78 https://jira.mongodb.org/browse/SERVER-6226
79 https://jira.mongodb.org/browse/SERVER-6961

612

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

ObjectId().toString() Returns String Literal ObjectId("...") In version 2.2, the toString()

method returns the string representation of the ObjectId() (page 129) object and has the format ObjectId("...").
Consider
the
following
example
that
calls
ObjectId("507c7f79bcf86cd7994f6c0e") object:

the

toString()

method

on

the

ObjectId("507c7f79bcf86cd7994f6c0e").toString()

The method now returns the string ObjectId("507c7f79bcf86cd7994f6c0e").

Previously, in version 2.0, the method would return the hexadecimal string 507c7f79bcf86cd7994f6c0e.
If compatibility between versions 2.0 and 2.2 is required, use ObjectId().str (page 129), which holds the hexadecimal
string value in both versions.
ObjectId().valueOf() Returns hexadecimal string In version 2.2, the valueOf() method returns the
value of the ObjectId() (page 129) object as a lowercase hexadecimal string.
Consider
the
following
example
that
calls
ObjectId("507c7f79bcf86cd7994f6c0e") object:

the

valueOf()

method

on

the

ObjectId("507c7f79bcf86cd7994f6c0e").valueOf()

The method now returns the hexadecimal string 507c7f79bcf86cd7994f6c0e.

Previously, in version 2.0, the method would return the object ObjectId("507c7f79bcf86cd7994f6c0e").
If compatibility between versions 2.0 and 2.2 is required, use ObjectId().str (page 129) attribute, which holds the
hexadecimal string value in both versions.
Behavioral Changes

Restrictions on Collection Names In version 2.2, collection names cannot:

contain the $.
be an empty string (i.e. "").
This change does not affect collections created with now illegal names in earlier versions of MongoDB.
These new restrictions are in addition to the existing restrictions on collection names which are:
A collection name should begin with a letter or an underscore.
A collection name cannot contain the null character.
Begin with the system. prefix.
system.indexes collection.

MongoDB reserves system.

for system collections, such as the

The maximum size of a collection name is 128 characters, including the name of the database. However, for
maximum flexibility, collections should have names less than 80 characters.
Collections names may have any other valid UTF-8 string.
See the SERVER-444280 and the Are there any restrictions on the names of Collections? (page 565) FAQ item.
80 https://jira.mongodb.org/browse/SERVER-4442

11.2. Previous Stable Releases

613

MongoDB Documentation, Release 2.4.9

Restrictions on Database Names for Windows Database names running on Windows can no longer contain the
following characters:
/\. "*<>:|?

The names of the data files include the database name. If you attempt to upgrade a database instance with one or more
of these characters, mongod will refuse to start.
Change the name of these databases before upgrading. See SERVER-458481 and SERVER-672982 for more information.
_id Fields and Indexes on Capped Collections All capped collections now have an _id field by default, if they
exist outside of the local database, and now have indexes on the _id field. This change only affects capped
collections created with 2.2 instances and does not affect existing capped collections.
See: SERVER-551683 for more information.
New $elemMatch Projection Operator The $elemMatch operator allows applications to narrow the data
returned from queries so that the query operation will only return the first matching element in an array.
See the http://docs.mongodb.org/manualreference/operator/projection/elemMatch documentation and the SERVER-223884 and SERVER-82885 issues for more information.
Windows Specific Changes

Windows XP is Not Supported As of 2.2, MongoDB does not support Windows XP. Please upgrade to a more
recent version of Windows to use the latest releases of MongoDB. See SERVER-564886 for more information.
Service Support for mongos.exe You may now run mongos.exe instances as a Windows Service. See the
http://docs.mongodb.org/manualreference/program/mongos.exe reference and MongoDB as a
Windows Service (page 15) and SERVER-158987 for more information.
Log Rotate Command Support MongoDB for Windows now supports log rotation by way of the logRotate
database command. See SERVER-261288 for more information.
New Build Using SlimReadWrite Locks for Windows Concurrency Labeled 2008+ on the Downloads Page89 ,
this build for 64-bit versions of Windows Server 2008 R2 and for Windows 7 or newer, offers increased performance
over the standard 64-bit Windows build of MongoDB. See SERVER-384490 for more information.
81 https://jira.mongodb.org/browse/SERVER-4584
82 https://jira.mongodb.org/browse/SERVER-6729
83 https://jira.mongodb.org/browse/SERVER-5516
84 https://jira.mongodb.org/browse/SERVER-2238
85 https://jira.mongodb.org/browse/SERVER-828
86 https://jira.mongodb.org/browse/SERVER-5648
87 https://jira.mongodb.org/browse/SERVER-1589
88 https://jira.mongodb.org/browse/SERVER-2612
89 http://www.mongodb.org/downloads
90 https://jira.mongodb.org/browse/SERVER-3844

614

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

Tool Improvements

Index Definitions Handled by mongodump and mongorestore When you specify the --collection option
to mongodump, mongodump will now backup the definitions for all indexes that exist on the source database. When
you attempt to restore this backup with mongorestore, the target mongod will rebuild all indexes. See SERVER80891 for more information.
mongorestore now includes the --noIndexRestore option to provide the preceding behavior.
--noIndexRestore to prevent mongorestore from building previous indexes.

Use

mongooplog for Replaying Oplogs The mongooplog tool makes it possible to pull oplog entries
from mongod instance and apply them to another mongod instance.
You can use mongooplog
to achieve point-in-time backup of a MongoDB data set.
See the SERVER-387392 case and the
http://docs.mongodb.org/manualreference/program/mongooplog documentation.
Authentication Support for mongotop and mongostat mongotop and mongostat now contain support for
username/password authentication. See SERVER-387593 and SERVER-387194 for more information regarding this
change. Also consider the documentation of the following options for additional information:
mongotop --username
mongotop --password
mongostat --username
mongostat --password
Write Concern Support for mongoimport and mongorestore mongoimport now provides an option to
halt the import if the operation encounters an error, such as a network interruption, a duplicate key exception, or a
write error. The --stopOnError option will produce an error rather than silently continue importing data. See
SERVER-393795 for more information.
In mongorestore, the --w option provides support for configurable write concern.
mongodump Support for Reading from Secondaries You can now run mongodump when connected to a secondary member of a replica set. See SERVER-385496 for more information.
mongoimport Support for full 16MB Documents Previously, mongoimport would only import documents
that were less than 4 megabytes in size. This issue is now corrected, and you may use mongoimport to import
documents that are at least 16 megabytes ins size. See SERVER-459397 for more information.
Timestamp() Extended JSON format MongoDB extended JSON now includes a new Timestamp() type to
represent the Timestamp type that MongoDB uses for timestamps in the oplog among other contexts.
This permits tools like mongooplog and mongodump to query for specific timestamps. Consider the following
mongodump operation:
91 https://jira.mongodb.org/browse/SERVER-808
92 https://jira.mongodb.org/browse/SERVER-3873
93 https://jira.mongodb.org/browse/SERVER-3875
94 https://jira.mongodb.org/browse/SERVER-3871
95 https://jira.mongodb.org/browse/SERVER-3937
96 https://jira.mongodb.org/browse/SERVER-3854
97 https://jira.mongodb.org/browse/SERVER-4593

11.2. Previous Stable Releases

615

MongoDB Documentation, Release 2.4.9

mongodump --db local --collection oplog.rs --query '{"ts":{"$gt":{"$timestamp" : {"t": 1344969612000,

See SERVER-348398 for more information.

Shell Improvements

Improved Shell User Interface 2.2 includes a number of changes that improve the overall quality and consistency
of the user interface for the mongo shell:
Full Unicode support.
Bash-like line editing features. See SERVER-431299 for more information.
Multi-line command support in shell history. See SERVER-3470100 for more information.
Windows support for the edit command. See SERVER-3998101 for more information.
Helper to load Server-Side Functions The db.loadServerScripts() loads the contents of the current
databases system.js collection into the current mongo shell session. See SERVER-1651102 for more information.
Support for Bulk Inserts If you pass an array of documents to the insert() method, the mongo shell will now
perform a bulk insert operation. See SERVER-3819103 and SERVER-2395104 for more information.
Note: For bulk inserts on sharded clusters, the getLastError command alone is insufficient to verify success.
Applications should must verify the success of bulk inserts in application logic.

Operations

Support for Logging to Syslog See the SERVER-2957105 case and the documentation of the syslog run-time
option or the mongod --syslog and mongos --syslog command line-options.
touch Command Added the touch command to read the data and/or indexes from a collection into memory. See:
SERVER-2023106 and touch for more information.
indexCounters No Longer Report Sampled Data indexCounters now report actual counters that reflect
index use and state. In previous versions, these data were sampled. See SERVER-5784107 and indexCounters for
more information.
Padding Specifiable on compact Command See the documentation of the compact and the SERVER-4018108
issue for more information.
98 https://jira.mongodb.org/browse/SERVER-3483
99 https://jira.mongodb.org/browse/SERVER-4312
100 https://jira.mongodb.org/browse/SERVER-3470
101 https://jira.mongodb.org/browse/SERVER-3998
102 https://jira.mongodb.org/browse/SERVER-1651
103 https://jira.mongodb.org/browse/SERVER-3819
104 https://jira.mongodb.org/browse/SERVER-2395
105 https://jira.mongodb.org/browse/SERVER-2957
106 https://jira.mongodb.org/browse/SERVER-2023
107 https://jira.mongodb.org/browse/SERVER-5784
108 https://jira.mongodb.org/browse/SERVER-4018

616

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

Added Build Flag to Use System Libraries The Boost library, version 1.49, is now embedded in the MongoDB
code base.
If you want to build MongoDB binaries using system Boost libraries, you can pass scons using the
--use-system-boost flag, as follows:
scons --use-system-boost

When building MongoDB, you can also pass scons a flag to compile MongoDB using only system libraries rather
than the included versions of the libraries. For example:
scons --use-system-all

See the SERVER-3829109 and SERVER-5172110 issues for more information.

Memory Allocator Changed to TCMalloc To improve performance, MongoDB 2.2 uses the TCMalloc memory
allocator from Google Perftools. For more information about this change see the SERVER-188111 and SERVER4683112 . For more information about TCMalloc, see the documentation of TCMalloc113 itself.
Replication

Improved Logging for Replica Set Lag When secondary members of a replica set fall behind in replication,
mongod now provides better reporting in the log. This makes it possible to track replication in general and identify what process may produce errors or halt replication. See SERVER-3575114 for more information.
Replica Set Members can Sync from Specific Members The new replSetSyncFrom command and new
rs.syncFrom() helper in the mongo shell make it possible for you to manually configure from which member of the set a replica will poll oplog entries. Use these commands to override the default selection logic if needed.
Always exercise caution with replSetSyncFrom when overriding the default behavior.
Replica Set Members will not Sync from Members Without Indexes Unless buildIndexes: false To
prevent inconsistency between members of replica sets, if the member of a replica set has buildIndexes (page 468)
set to true, other members of the replica set will not sync from this member, unless they also have buildIndexes
(page 468) set to true. See SERVER-4160115 for more information.
New Option To Configure Index Pre-Fetching during Replication By default, when replicating options, secondaries will pre-fetch Indexes (page 313) associated with a query to improve replication throughput in most cases. The
replIndexPrefetch setting and --replIndexPrefetch option allow administrators to disable this feature
or allow the mongod to pre-fetch only the index on the _id field. See SERVER-6718116 for more information.
Map Reduce Improvements

In 2.2 Map Reduce received the following improvements:

109 https://jira.mongodb.org/browse/SERVER-3829
110 https://jira.mongodb.org/browse/SERVER-5172
111 https://jira.mongodb.org/browse/SERVER-188
112 https://jira.mongodb.org/browse/SERVER-4683
113 http://goog-perftools.sourceforge.net/doc/tcmalloc.html
114 https://jira.mongodb.org/browse/SERVER-3575
115 https://jira.mongodb.org/browse/SERVER-4160
116 https://jira.mongodb.org/browse/SERVER-6718

11.2. Previous Stable Releases

617

MongoDB Documentation, Release 2.4.9

Improved support for sharded MapReduce117 , and

MapReduce will retry jobs following a config error118 .
Sharding Improvements

Index on Shard Keys Can Now Be a Compound Index If your shard key uses the prefix of an existing index,
then you do not need to maintain a separate index for your shard key in addition to your existing index. This index,
however, cannot be a multi-key index. See the Shard Key Indexes (page 505) documentation and SERVER-1506119
for more information.
Migration Thresholds Modified The migration thresholds (page 502) have changed in 2.2 to permit more even
distribution of chunks in collections that have smaller quantities of data. See the Migration Thresholds (page 502)
documentation for more information.
Licensing Changes
Added License notice for Google Perftools (TCMalloc Utility). See the License Notice120 and the SERVER-4683121
for more information.
Resources
MongoDB Downloads122 .
All JIRA issues resolved in 2.2123 .
All backwards incompatible changes124 .
All third party license notices125 .
Whats New in MongoDB 2.2 Online Conference126 .

11.2.2 Release Notes for MongoDB 2.0

Upgrading
Although the major version number has changed, MongoDB 2.0 is a standard, incremental production release and
works as a drop-in replacement for MongoDB 1.8.
117 https://jira.mongodb.org/browse/SERVER-4521
118 https://jira.mongodb.org/browse/SERVER-4158
119 https://jira.mongodb.org/browse/SERVER-1506
120 https://github.com/mongodb/mongo/blob/v2.2/distsrc/THIRD-PARTY-NOTICES#L231
121 https://jira.mongodb.org/browse/SERVER-4683
122 http://mongodb.org/downloads

123 https://jira.mongodb.org/secure/IssueNavigator.jspa?reset=true&jqlQuery=project+%3D+SERVER+AND+fixVersion+in+%28%222.1.0%22%2C+%222.1.1%22%2
rc0%22%2C+%222.2.0-rc1%22%2C+%222.2.0-rc2%22%29+ORDER+BY+component+ASC%2C+key+DESC
124 https://jira.mongodb.org/secure/IssueNavigator.jspa?requestId=11225
125 https://github.com/mongodb/mongo/blob/v2.2/distsrc/THIRD-PARTY-NOTICES
126 http://www.mongodb.com/events/webinar/mongodb-online-conference-sept

618

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

Preparation

Read through all release notes before upgrading, and ensure that no changes will affect your deployment.
If you create new indexes in 2.0, then downgrading to 1.8 is possible but you must reindex the new collections.
mongoimport and mongoexport now correctly adhere to the CSV spec for handling CSV input/output. This
may break existing import/export workflows that relied on the previous behavior. For more information see SERVER1097127 .
Journaling128 is enabled by default in 2.0 for 64-bit builds. If you still prefer to run without journaling, start mongod
with the --nojournal run-time option. Otherwise, MongoDB creates journal files during startup. The first time
you start mongod with journaling, you will see a delay as mongod creates new files. In addition, you may see reduced
write throughput.
2.0 mongod instances are interoperable with 1.8 mongod instances; however, for best results, upgrade your deployments using the following procedures:
Upgrading a Standalone mongod

1. Download the v2.0.x binaries from the MongoDB Download Page129 .

2. Shutdown your mongod instance. Replace the existing binary with the 2.0.x mongod binary and restart MongoDB.
Upgrading a Replica Set

1. Upgrade the secondary members of the set one at a time by shutting down the mongod and replacing the 1.8
binary with the 2.0.x binary from the MongoDB Download Page130 .
2. To avoid losing the last few updates on failover you can temporarily halt your application (failover should take
less than 10 seconds), or you can set write concern (page 47) in your application code to confirm that each
update reaches multiple servers.
3. Use the rs.stepDown() to step down the primary to allow the normal failover (page 397) procedure.
rs.stepDown() and replSetStepDown provide for shorter and more consistent failover procedures than
simply shutting down the primary directly.
When the primary has stepped down, shut down its instance and upgrade by replacing the mongod binary with
the 2.0.x binary.
Upgrading a Sharded Cluster

1. Upgrade all config server instances first, in any order. Since config servers use two-phase commit, shard configuration metadata updates will halt until all are up and running.
2. Upgrade mongos routers in any order.
127 https://jira.mongodb.org/browse/SERVER-1097
128 http://www.mongodb.org/display/DOCS/Journaling
129 http://downloads.mongodb.org/
130 http://downloads.mongodb.org/

11.2. Previous Stable Releases

619

MongoDB Documentation, Release 2.4.9

Changes
Compact Command

A compact command is now available for compacting a single collection and its indexes. Previously, the only way
to compact was to repair the entire database.
Concurrency Improvements

When going to disk, the server will yield the write lock when writing data that is not likely to be in memory. The
initial implementation of this feature now exists:
See SERVER-2563131 for more information.
The specific operations yield in 2.0 are:
Updates by _id
Removes
Long cursor iterations
Default Stack Size

MongoDB 2.0 reduces the default stack size. This change can reduce total memory usage when there are many (e.g.,
1000+) client connections, as there is a thread per connection. While portions of a threads stack can be swapped out
if unused, some operating systems do this slowly enough that it might be an issue. The default stack size is lesser of
the system setting or 1MB.
Index Performance Enhancements

v2.0 includes significant improvements to the index (page 346). Indexes are often 25% smaller and 25% faster (depends
on the use case). When upgrading from previous versions, the benefits of the new index type are realized only if you
create a new index or re-index an old one.
Dates are now signed, and the max index key size has increased slightly from 819 to 1024 bytes.
All operations that create a new index will result in a 2.0 index by default. For example:
Reindexing results on an older-version index results in a 2.0 index. However, reindexing on a secondary does
not work in versions prior to 2.0. Do not reindex on a secondary. For a workaround, see SERVER-3866132 .
The repairDatabase command converts indexes to a 2.0 indexes.
To convert all indexes for a given collection to the 2.0 type (page 620), invoke the compact command.
Once you create new indexes, downgrading to 1.8.x will require a re-index of any indexes created using 2.0. See Build
Old Style Indexes (page 346).
Sharding Authentication

Applications can now use authentication with sharded clusters.

131 https://jira.mongodb.org/browse/SERVER-2563
132 https://jira.mongodb.org/browse/SERVER-3866

620

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

Replica Sets

Hidden Nodes in Sharded Clusters In 2.0, mongos instances can now determine when a member of a replica set
becomes hidden without requiring a restart. In 1.8, mongos if you reconfigured a member as hidden, you had to
restart mongos to prevent queries from reaching the hidden member.
Priorities Each replica set member can now have a priority value consisting of a floating-point from 0 to 1000,
inclusive. Priorities let you control which member of the set you prefer to have as primary the member with the
highest priority that can see a majority of the set will be elected primary.
For example, suppose you have a replica set with three members, A, B, and C, and suppose that their priorities are set
as follows:
As priority is 2.
Bs priority is 3.
Cs priority is 1.
During normal operation, the set will always chose B as primary. If B becomes unavailable, the set will elect A as
primary.
For more information, see the priority (page 469) documentation.
Data-Center Awareness You can now tag replica set members to indicate their location. You can use these tags
to design custom write rules (page 47) across data centers, racks, specific servers, or any other architecture choice.
For example, an administrator can define rules such as very important write or customerData or audit-trail to
replicate to certain servers, racks, data centers, etc. Then in the application code, the developer would say:
db.foo.insert(doc, {w : "very important write"})

which would succeed if it fulfilled the conditions the DBA defined for very important write.
For more information, see Tagging133 .
Drivers may also support tag-aware reads. Instead of specifying slaveOk, you specify slaveOk with tags indicating
which data-centers to read from. For details, see the MongoDB Drivers and Client Libraries (page 95) documentation.
w : majority You can also set w to majority to ensure that the write propagates to a majority of nodes, effectively committing it. The value for majority will automatically adjust as you add or remove nodes from the
set.
For more information, see Write Concern (page 47).
Reconfiguration with a Minority Up If the majority of servers in a set has been permanently lost, you can now
force a reconfiguration of the set to bring it back online.
For more information see Reconfigure a Replica Set with Unavailable Members (page 454).
Primary Checks for a Caught up Secondary before Stepping Down To minimize time without a primary, the
rs.stepDown() method will now fail if the primary does not see a secondary within 10 seconds of its latest
optime. You can force the primary to step down anyway, but by default it will return an error message.
See also Force a Member to Become Primary (page 447).
133 http://www.mongodb.org/display/DOCS/Data+Center+Awareness#DataCenterAwareness-Tagging%28version2.0%29

11.2. Previous Stable Releases

621

MongoDB Documentation, Release 2.4.9

Extended Shutdown on the Primary to Minimize Interruption When you call the shutdown command, the
primary will refuse to shut down unless there is a secondary whose optime is within 10 seconds of the primary. If such
a secondary isnt available, the primary will step down and wait up to a minute for the secondary to be fully caught up
before shutting down.
Note that to get this behavior, you must issue the shutdown command explicitly; sending a signal to the process will
not trigger this behavior.
You can also force the primary to shut down, even without an up-to-date secondary available.
Maintenance Mode When repair or compact runs on a secondary, the secondary will automatically drop into
recovering mode until the operation finishes. This prevents clients from trying to read from it while its busy.
Geospatial Features

Multi-Location Documents Indexing is now supported on documents which have multiple location objects, embedded either inline or in nested sub-documents. Additional command options are also supported, allowing results to
return with not only distance but the location used to generate the distance.
For more information, see Multi-location Documents134 .
Polygon searches Polygonal $within queries are also now supported for simple polygon shapes. For details, see
the $within operator documentation.
Journaling Enhancements

Journaling is now enabled by default for 64-bit platforms. Use the --nojournal command line option to
disable it.
The journal is now compressed for faster commits to disk.
A new --journalCommitInterval run-time option exists for specifying your own group commit interval.
The default settings do not change.
A new { getLastError: { j: true } } option is available to wait for the group commit. The
group commit will happen sooner when a client is waiting on {j: true}. If journaling is disabled, {j:
true} is a no-op.
New ContinueOnError Option for Bulk Insert

Set the continueOnError option for bulk inserts, in the driver (page 95), so that bulk insert will continue to insert
any remaining documents even if an insert fails, as is the case with duplicate key exceptions or network interruptions.
The getLastError command will report whether any inserts have failed, not just the last one. If multiple errors
occur, the client will only receive the most recent getLastError results.
See OP_INSERT135 .
Note: For bulk inserts on sharded clusters, the getLastError command alone is insufficient to verify success.
Applications should must verify the success of bulk inserts in application logic.
134 http://www.mongodb.org/display/DOCS/Geospatial+Indexing#GeospatialIndexing-MultilocationDocuments
135 http://www.mongodb.org/display/DOCS/Mongo+Wire+Protocol#MongoWireProtocol-OPINSERT

622

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

Map Reduce

Output to a Sharded Collection Using the new sharded flag, it is possible to send the result of a map/reduce to
a sharded collection. Combined with the reduce or merge flags, it is possible to keep adding data to very large
collections from map/reduce jobs.

For more information, see MapReduce Output Options136 and http://docs.mongodb.org/manualreference/command/ma

Performance Improvements Map/reduce performance will benefit from the following:
Larger in-memory buffer sizes, reducing the amount of disk I/O needed during a job
Larger javascript heap size, allowing for larger objects and less GC
Supports
pure
JavaScript
execution
with
the
jsMode
flag.
http://docs.mongodb.org/manualreference/command/mapReduce.

See

New Querying Features

Additional regex options: s Allows the dot (.) to match all characters including new lines. This is in addition to
the currently supported i, m and x. See Regular Expressions137 and $regex.
$and A special boolean $and query operator is now available.
Command Output Changes

The output of the validate command and the documents in the system.profile collection have both been
enhanced to return information as BSON objects with keys for each value rather than as free-form strings.
Shell Features

Custom Prompt You can define a custom prompt for the mongo shell. You can change the prompt at any time by
setting the prompt variable to a string or a custom JavaScript function returning a string. For examples, see Custom
Prompt138 .
Default Shell Init Script On startup, the shell will check for a .mongorc.js file in the users home directory.
The shell will execute this file after connecting to the database and before displaying the prompt.
If you would like the shell not to run the .mongorc.js file automatically, start the shell with --norc.
For more information, see http://docs.mongodb.org/manualreference/program/mongo.
Most Commands Require Authentication

In 2.0, when running with authentication (e.g. auth) all database commands require authentication, except the following commands.
isMaster
136 http://www.mongodb.org/display/DOCS/MapReduce#MapReduce-Outputoptions
137 http://www.mongodb.org/display/DOCS/Advanced+Queries#AdvancedQueries-RegularExpressions
138 http://www.mongodb.org/display/DOCS/Overview+-+The+MongoDB+Interactive+Shell#Overview-TheMongoDBInteractiveShellCustomPrompt

11.2. Previous Stable Releases

623

MongoDB Documentation, Release 2.4.9

authenticate
getnonce
buildInfo
ping
isdbgrid
Resources
MongoDB Downloads139
All JIRA Issues resolved in 2.0140
All Backward Incompatible Changes141

11.2.3 Release Notes for MongoDB 1.8

Upgrading
MongoDB 1.8 is a standard, incremental production release and works as a drop-in replacement for MongoDB 1.6,
except:
Replica set members should be upgraded in a particular order, as described in Upgrading a Replica Set
(page 625).
The mapReduce command has changed in 1.8, causing incompatibility with previous releases. mapReduce
no longer generates temporary collections (thus, keepTemp has been removed). Now, you must always supply
a value for out. See the out field options in the mapReduce document. If you use MapReduce, this also
likely means you need a recent version of your client driver.
Preparation

Read through all release notes before upgrading and ensure that no changes will affect your deployment.
Upgrading a Standalone mongod

1. Download the v1.8.x binaries from the MongoDB Download Page142 .

2. Shutdown your mongod instance.
3. Replace the existing binary with the 1.8.x mongod binary.
4. Restart MongoDB.
139 http://mongodb.org/downloads
140 https://jira.mongodb.org/secure/IssueNavigator.jspa?mode=hide&requestId=11002
141 https://jira.mongodb.org/secure/IssueNavigator.jspa?requestId=11023
142 http://downloads.mongodb.org/

624

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

Upgrading a Replica Set

1.8.x secondaries can replicate from 1.6.x primaries.

1.6.x secondaries cannot replicate from 1.8.x primaries.
Thus, to upgrade a replica set you must replace all of your secondaries first, then the primary.
For example, suppose you have a replica set with a primary, an arbiter and several secondaries. To upgrade the set, do
the following:
1. For the arbiter:
(a) Shut down the arbiter.
(b) Restart it with the 1.8.x binary from the MongoDB Download Page143 .
2. Change your config (optional) to prevent election of a new primary.
It is possible that, when you start shutting down members of the set, a new primary will be elected. To prevent
this, you can give all of the secondaries a priority of 0 before upgrading, and then change them back afterwards.
To do so:
(a) Record your current config. Run rs.config() and paste the results into a text file.
(b) Update your config so that all secondaries have priority 0. For example:
config = rs.conf()
{
"_id" : "foo",
"version" : 3,
"members" : [
{
"_id" : 0,
"host" : "ubuntu:27017"
},
{
"_id" : 1,
"host" : "ubuntu:27018"
},
{
"_id" : 2,
"host" : "ubuntu:27019",
"arbiterOnly" : true
}
{
"_id" : 3,
"host" : "ubuntu:27020"
},
{
"_id" : 4,
"host" : "ubuntu:27021"
},
]
}
config.version++
3
rs.isMaster()
{
"setName" : "foo",
143 http://downloads.mongodb.org/

11.2. Previous Stable Releases

625

MongoDB Documentation, Release 2.4.9

"ismaster" : false,
"secondary" : true,
"hosts" : [
"ubuntu:27017",
"ubuntu:27018"
],
"arbiters" : [
"ubuntu:27019"
],
"primary" : "ubuntu:27018",
"ok" : 1
}
// for each secondary
config.members[0].priority = 0
config.members[3].priority = 0
config.members[4].priority = 0
rs.reconfig(config)

3. For each secondary:

(a) Shut down the secondary.
(b) Restart it with the 1.8.x binary from the MongoDB Download Page144 .
4. If you changed the config, change it back to its original state:
config = rs.conf()
config.version++
config.members[0].priority = 1
config.members[3].priority = 1
config.members[4].priority = 1
rs.reconfig(config)

5. Shut down the primary (the final 1.6 server), and then restart it with the 1.8.x binary from the MongoDB
Download Page145 .
Upgrading a Sharded Cluster

1. Turn off the balancer:

mongo <a_mongos_hostname>
use config
db.settings.update({_id:"balancer"},{$set : {stopped:true}}, true)

2. For each shard:

If the shard is a replica set, follow the directions above for Upgrading a Replica Set (page 625).
If the shard is a single mongod process, shut it down and then restart it with the 1.8.x binary from the
MongoDB Download Page146 .
3. For each mongos:
(a) Shut down the mongos process.
(b) Restart it with the 1.8.x binary from the MongoDB Download Page147 .
144 http://downloads.mongodb.org/
145 http://downloads.mongodb.org/
146 http://downloads.mongodb.org/
147 http://downloads.mongodb.org/

626

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

4. For each config server:

(a) Shut down the config server process.
(b) Restart it with the 1.8.x binary from the MongoDB Download Page148 .
5. Turn on the balancer:
use config
db.settings.update({_id:"balancer"},{$set : {stopped:false}})

Returning to 1.6

If for any reason you must move back to 1.6, follow the steps above in reverse. Please be careful that you have not
inserted any documents larger than 4MB while running on 1.8 (where the max size has increased to 16MB). If you
have you will get errors when the server tries to read those documents.
Journaling Returning to 1.6 after using 1.8 Journaling (page 234) works fine, as journaling does not change anything
about the data file format. Suppose you are running 1.8.x with journaling enabled and you decide to switch back to
1.6. There are two scenarios:
If you shut down cleanly with 1.8.x, just restart with the 1.6 mongod binary.
If 1.8.x shut down uncleanly, start 1.8.x up again and let the journal files run to fix any damage (incomplete
writes) that may have existed at the crash. Then shut down 1.8.x cleanly and restart with the 1.6 mongod binary.
Changes
Journaling

MongoDB now supports write-ahead Journaling Mechanics (page 234) to facilitate fast crash recovery and durability
in the storage engine. With journaling enabled, a mongod can be quickly restarted following a crash without needing
to repair the collections. The aggregation framework makes it possible to do aggregation
Sparse and Covered Indexes

Sparse Indexes (page 335) are indexes that only include documents that contain the fields specified in the index.
Documents missing the field will not appear in the index at all. This can significantly reduce index size for indexes of
fields that contain only a subset of documents within a collection.
Covered Indexes (page 314) enable MongoDB to answer queries entirely from the index when the query only selects
fields that the index contains.
Incremental MapReduce Support

The mapReduce command supports new options that enable incrementally updating existing collections. Previously,
a MapReduce job could output either to a temporary collection or to a named permanent collection, which it would
overwrite with new data.
You now have several options for the output of your MapReduce jobs:
148 http://downloads.mongodb.org/

11.2. Previous Stable Releases

627

MongoDB Documentation, Release 2.4.9

You can merge MapReduce output into an existing collection. Output from the Reduce phase will replace
existing keys in the output collection if it already exists. Other keys will remain in the collection.
You can now re-reduce your output with the contents of an existing collection. Each key output by the reduce
phase will be reduced with the existing document in the output collection.
You can replace the existing output collection with the new results of the MapReduce job (equivalent to setting
a permanent output collection in previous releases)
You can compute MapReduce inline and return results to the caller without persisting the results of the job. This
is similar to the temporary collections generated in previous releases, except results are limited to 8MB.
For more information, see the out field options in the mapReduce document.
Additional Changes and Enhancements

1.8.1
Sharding migrate fix when moving larger chunks.
Durability fix with background indexing.
Fixed mongos concurrency issue with many incoming connections.
1.8.0
All changes from 1.7.x series.
1.7.6
Bug fixes.
1.7.5
Journaling (page 234).
Extent allocation improvements.
Improved replica set connectivity for mongos.
getLastError improvements for sharding.
1.7.4
mongos routes slaveOk queries to secondaries in replica sets.
New mapReduce output options.
Sparse Indexes (page 335).
1.7.3
Initial covered index (page 314) support.
Distinct can use data from indexes when possible.
mapReduce can merge or reduce results into an existing collection.
mongod tracks and mongostat displays network usage. See mongostat.

628

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

Sharding stability improvements.

1.7.2
$rename operator allows renaming of fields in a document.
db.eval() not to block.
Geo queries with sharding.
mongostat --discover option
Chunk splitting enhancements.
Replica sets network enhancements for servers behind a nat.
1.7.1
Many sharding performance enhancements.
Better support for $elemMatch on primitives in embedded arrays.
Query optimizer enhancements on range queries.
Window service enhancements.
Replica set setup improvements.
$pull works on primitives in arrays.
1.7.0
Sharding performance improvements for heavy insert loads.
Slave delay support for replica sets.
getLastErrorDefaults (page 470) for replica sets.
Auto completion in the shell.
Spherical distance for geo search.
All fixes from 1.6.1 and 1.6.2.
Release Announcement Forum Pages

1.8.1149 , 1.8.0150
1.7.6151 , 1.7.5152 , 1.7.4153 , 1.7.3154 , 1.7.2155 , 1.7.1156 , 1.7.0157
149 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/v09MbhEm62Y
150 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/JeHQOnam6Qk
151 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/3t6GNZ1qGYc
152 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/S5R0Tx9wkEg
153 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/9Om3Vuw-y9c
154 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/DfNUrdbmflI
155 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/df7mwK6Xixo
156 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/HUR9zYtTpA8
157 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/TUnJCg9161A

11.2. Previous Stable Releases

629

MongoDB Documentation, Release 2.4.9

Resources
MongoDB Downloads158
All JIRA Issues resolved in 1.8159

11.2.4 Release Notes for MongoDB 1.6

Upgrading
MongoDB 1.6 is a drop-in replacement for 1.4. To upgrade, simply shutdown mongod then restart with the new
binaries.
Please note that you should upgrade to the latest version of whichever driver youre using. Certain drivers, including
the Ruby driver, will require the upgrade, and all the drivers will provide extra features for connecting to replica sets.
Sharding
Sharding (page 479) is now production-ready, making MongoDB horizontally scalable, with no single point of failure.
A single instance of mongod can now be upgraded to a distributed cluster with zero downtime when the need arises.
Sharding (page 479)
Deploy a Sharded Cluster (page 507)
Convert a Replica Set to a Replicated Sharded Cluster (page 515)
Replica Sets
Replica sets (page 377), which provide automated failover among a cluster of n nodes, are also now available.
Please note that replica pairs are now deprecated; we strongly recommend that replica pair users upgrade to replica
sets.
Replication (page 377)
Deploy a Replica Set (page 420)
Convert a Standalone to a Replica Set (page 432)
Other Improvements
The w option (and wtimeout) forces writes to be propagated to n servers before returning success (this works
especially well with replica sets)
$or queries
Improved concurrency
$slice operator for returning subsets of arrays
64 indexes per collection (formerly 40 indexes per collection)
64-bit integers can now be represented in the shell using NumberLong
The findAndModify command now supports upserts. It also allows you to specify fields to return
158 http://mongodb.org/downloads
159 https://jira.mongodb.org/secure/IssueNavigator.jspa?mode=hide&requestId=10172

630

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

$showDiskLoc option to see disk location of a document

Support for IPv6 and UNIX domain sockets
Installation
Windows service improvements
The C++ client is a separate tarball from the binaries
1.6.x Release Notes
1.6.5160
1.5.x Release Notes
1.5.8161
1.5.7162
1.5.6163
1.5.5164
1.5.4165
1.5.3166
1.5.2167
1.5.1168
1.5.0169
You can see a full list of all changes on JIRA170 .
Thank you everyone for your support and suggestions!

11.2.5 Release Notes for MongoDB 1.4

Upgrading
Were pleased to announce the 1.4 release of MongoDB. 1.4 is a drop-in replacement for 1.2. To upgrade you just
need to shutdown mongod, then restart with the new binaries. (Users upgrading from release 1.0 should review the
1.2 release notes (page 633), in particular the instructions for upgrading the DB format.)
Release 1.4 includes the following improvements over release 1.2:
160 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/06_QCC05Fpk
161 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/uJfF1QN6Thk
162 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/OYvz40RWs90
163 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/4l0N2U_H0cQ
164 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/oO749nvTARY
165 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/380V_Ec_q1c
166 https://groups.google.com/forum/?hl=en&fromgroups=#!topic/mongodb-user/hsUQL9CxTQw
167 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/94EE3HVidAA
168 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/7SBPQ2RSfdM
169 https://groups.google.com/forum/?fromgroups=#!topic/mongodb-user/VAhJcjDGTy0
170 https://jira.mongodb.org/secure/IssueNavigator.jspa?mode=hide&requestId=10107

11.2. Previous Stable Releases

631

MongoDB Documentation, Release 2.4.9

Core Server Enhancements

concurrency (page 569) improvements
indexing memory improvements
background index creation (page 336)
better detection of regular expressions so the index can be used in more cases
Replication and Sharding
better handling for restarting slaves offline for a while
fast new slaves from snapshots (--fastsync)
configurable slave delay (--slavedelay)
replication handles clock skew on master
$inc replication fixes
sharding alpha 3 - notably 2-phase commit on config servers
Deployment and Production
configure slow threshold for profiling (page 174)
ability to do fsync + lock for backing up raw files
option for separate directory per db (--directoryperdb)
http://localhost:28017/_status to get serverStatus via http
REST interface is off by default for security (--rest to enable)
can rotate logs with a db command, logRotate
enhancements to serverStatus command (db.serverStatus()) - counters and replication lag (page 461) stats
new mongostat tool
Query Language Improvements
$all with regex
$not
partial matching of array elements $elemMatch
$ operator for updating arrays
$addToSet
$unset
$pull supports object matching
$set with array indexes

632

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

Geo
2d geospatial search (page 331)
geo $center and $box searches

11.2.6 Release Notes for MongoDB 1.2.x

New Features
More indexes per collection
Faster index creation
Map/Reduce
Stored JavaScript functions
Configurable fsync time
Several small features and fixes
DB Upgrade Required
There are some changes that will require doing an upgrade if your previous version is <= 1.0.x. If youre already using
a version >= 1.1.x then these changes arent required. There are 2 ways to do it:
--upgrade
stop your mongod process
run ./mongod --upgrade
start mongod again
use a slave
start a slave on a different port and data directory
when its synced, shut down the master, and start the new slave on the regular port.
Ask in the forums or IRC for more help.
Replication Changes
There have been minor changes in replication. If you are upgrading a master/slave setup from <= 1.1.2 you have
to update the slave first.
mongoimport
mongoimportjson has been removed and is replaced with mongoimport that can do json/csv/tsv
field filter changing
Weve changed the semantics of the field filter a little bit. Previously only objects with those fields would be
returned. Now the field filter only changes the output, not which objects are returned. If you need that behavior,
you can use $exists

11.2. Previous Stable Releases

633

MongoDB Documentation, Release 2.4.9

11.3 Other MongoDB Release Notes

11.3.1 Default Write Concern Change
These release notes outline a change to all driver interfaces released in November 2012. See release notes for specific
drivers for additional information.
Changes
As of the releases listed below, there are two major changes to all drivers:
1. All drivers will add a new top-level connection class that will increase consistency for all MongoDB client
interfaces.
This change is non-backward breaking: existing connection classes will remain in all drivers for a time, and will
continue to operate as expected. However, those previous connection classes are now deprecated as of these
releases, and will eventually be removed from the driver interfaces.
The new top-level connection class is named MongoClient, or similar depending on how host languages
handle namespacing.
2. The default write concern on the new MongoClient class will be to acknowledge all write operations
This will allow your application to receive acknowledgment of all write operations.

171

See the documentation of Write Concern (page 47) for more information about write concern in MongoDB.
Please migrate to the new MongoClient class expeditiously.
Releases
The following driver releases will include the changes outlined in Changes (page 634). See each drivers release notes
for a full account of each release as well as other related driver-specific changes.
C#, version 1.7
Java, version 2.10.0
Node.js, version 1.2
Perl, version 0.501.1
PHP, version 1.4
Python, version 2.4
Ruby, version 1.8

11.4 MongoDB Version Numbers

For MongoDB 2.4.1, 2.4 refers to the release series and .1 refers to the revision. The second component of the
release series (e.g. 4 in 2.4.1) describes the type of release series. Release series ending with even numbers (e.g. 4
above) are stable and ready for production, while odd numbers are for development and testing only.
171 The drivers will call getLastError without arguments, which is logically equivalent to the w:
1 option; however, this operation allows
replica set users to override the default write concern with the getLastErrorDefaults (page 470) setting in the Replica Set Configuration
(page 467).

634

Chapter 11. Release Notes

MongoDB Documentation, Release 2.4.9

Generally, changes in the release series (e.g. 2.2 to 2.4) mark the introduction of new features that may break backwards compatibility. Changes to the revision number mark the release bug fixes and backwards-compatible changes.
Important: Always upgrade to the latest stable revision of your release series.
The version numbering system for MongoDB differs from the system used for the MongoDB drivers. Drivers use only
the first number to indicate a major version. For details, see Driver Version Numbers (page 96).
Example
Version numbers
2.0.0 : Stable release.
2.0.1 : Revision.
2.1.0 : Development release for testing only. Includes new features and changes for testing. Interfaces and
stability may not be compatible in development releases.
2.2.0 : Stable release. This is a culmination of the 2.1.x development series.

11.4. MongoDB Version Numbers

635

MongoDB Documentation, Release 2.4.9

636

Chapter 11. Release Notes

CHAPTER 12

About MongoDB Documentation

The MongoDB Manual1 contains comprehensive documentation on the MongoDB document-oriented database management system. This page describes the manuals licensing, editions, and versions, and describes how to make a
change request and how to contribute to the manual.
For more information on MongoDB, see MongoDB: A Document Oriented Database2 . To download MongoDB, see
the downloads page3 .

12.1 License
This manual is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported4 (i.e.
CC-BY-NC-SA) license.
The MongoDB Manual is copyright 2011-2014 MongoDB, Inc.

12.2 Editions
In addition to the MongoDB Manual5 , you can also access this content in the following editions:
ePub Format6
Single HTML Page7
PDF Format8 (without reference.)
HTML tar.gz9
You also can access PDF files that contain subsets of the MongoDB Manual:
MongoDB Reference Manual10
MongoDB CRUD Operations11
1 http://docs.mongodb.org/manual/#
2 http://www.mongodb.org/about/
3 http://www.mongodb.org/downloads
4 http://creativecommons.org/licenses/by-nc-sa/3.0/
5 http://docs.mongodb.org/manual/#
6 http://docs.mongodb.org/v2.4/MongoDB-manual.epub
7 http://docs.mongodb.org/v2.4/single/
8 http://docs.mongodb.org/v2.4/MongoDB-manual.pdf
9 http://docs.mongodb.org/v2.4/manual.tar.gz
10 http://docs.mongodb.org/v2.4/MongoDB-reference-manual.pdf
11 http://docs.mongodb.org/v2.4/MongoDB-crud-guide.pdf

637

MongoDB Documentation, Release 2.4.9

Data Models for MongoDB12

MongoDB Data Aggregation13
Replication and MongoDB14
Sharding and MongoDB15
MongoDB Administration16
MongoDB Security17
MongoDB Reference documentation is also available as part of dash18 . You can also access the MongoDB Man
Pages19 which are also distributed with the official MongoDB Packages.

12.3 Version and Revisions

This version of the manual reflects version 2.4 of MongoDB.
See the MongoDB Documentation Project Page20 for an overview of all editions and output formats of the MongoDB
Manual. You can see the full revision history and track ongoing improvements and additions for all versions of the
manual from its GitHub repository21 .
This edition reflects v2.4 branch of the documentation as of the 178ad486ae88fbfbeaa4961de340f5954070ff92
revision. This branch is explicitly accessible via http://docs.mongodb.org/v2.4 and you can always reference the
commit of the current manual in the release.txt22 file.
The most up-to-date, current,
http://docs.mongodb.org/manual/.

and

stable

version

of

the

manual

is

always

available

at

12.4 Report an Issue or Make a Change Request

To report an issue with this manual or to make a change request, file a ticket at the MongoDB DOCS Project on Jira23 .

12.5 Contribute to the Documentation

12.5.1 MongoDB Manual Translation
The original authorship language for all MongoDB documentation is American English. However, ensuring that
speakers of other languages can read and understand the documentation is of critical importance to the documentation
project.
12 http://docs.mongodb.org/v2.4/MongoDB-data-models-guide.pdf
13 http://docs.mongodb.org/v2.4/MongoDB-aggregation-guide.pdf
14 http://docs.mongodb.org/v2.4/MongoDB-replication-guide.pdf
15 http://docs.mongodb.org/v2.4/MongoDB-sharding-guide.pdf
16 http://docs.mongodb.org/v2.4/MongoDB-administration-guide.pdf
17 http://docs.mongodb.org/v2.4/MongoDB-security-guide.pdf
18 http://kapeli.com/dash
19 http://docs.mongodb.org/v2.4/manpages.tar.gz
20 http://docs.mongodb.org
21 https://github.com/mongodb/docs
22 http://docs.mongodb.org/v2.4/release.txt
23 https://jira.mongodb.org/browse/DOCS

638

Chapter 12. About MongoDB Documentation

MongoDB Documentation, Release 2.4.9

In this direction, the MongoDB Documentation project uses the service provided by Smartling24 to translate the MongoDB documentation into additional non-English languages. This translation project is largely supported by the work
of volunteer translators from the MongoDB community who contribute to the translation effort.
If you would like to volunteer to help translate the MongoDB documentation, please:
complete the MongoDB Contributor Agreement25 , and
create an account on Smartling at translate.docs.mongodb.org26 .
Please use the same email address you use to sign the contributor as you use to create your Smartling account.
The mongodb-translators27 user group exists to facilitate collaboration between translators and the documentation
team at large. You can join the Google Group without signing the contributors agreement.
We currently have the following languages configured:
Arabic28
Chinese29
Czech30
French31
German32
Hungarian33
Indonesian34
Italian35
Japanese36
Korean37
Lithuanian38
Polish39
Portuguese40
Romanian41
Russian42
Spanish43
24 http://smartling.com/
25 http://www.mongodb.com/legal/contributor-agreement
26 http://translate.docs.mongodb.org/
27 http://groups.google.com/group/mongodb-translators
28 http://ar.docs.mongodb.org
29 http://cn.docs.mongodb.org
30 http://cs.docs.mongodb.org
31 http://fr.docs.mongodb.org
32 http://de.docs.mongodb.org
33 http://hu.docs.mongodb.org
34 http://id.docs.mongodb.org
35 http://it.docs.mongodb.org
36 http://jp.docs.mongodb.org
37 http://ko.docs.mongodb.org
38 http://lt.docs.mongodb.org
39 http://pl.docs.mongodb.org
40 http://pt.docs.mongodb.org
41 http://ro.docs.mongodb.org
42 http://ru.docs.mongodb.org
43 http://es.docs.mongodb.org

12.5. Contribute to the Documentation

639

MongoDB Documentation, Release 2.4.9

Turkish44
Ukrainian45
If you would like to initiate a translation project to an additional language, please report this issue using the Report a
Problem link above or by posting to the mongodb-translators46 list.
Currently the translation project only publishes rendered translation. While the translation effort is currently focused
on the web site we are evaluating how to retrieve the translated phrases for use in other media.
See also:
Contribute to the Documentation (page 638)
Style Guide and Documentation Conventions (page 640)
MongoDB Manual Organization (page 649)
MongoDB Documentation Practices and Processes (page 646)
MongoDB Documentation Build System (page 650)
The entire documentation source for this manual is available in the mongodb/docs repository47 , which is one of the
MongoDB project repositories on GitHub48 .
To contribute to the documentation, you can open a GitHub account49 , fork the mongodb/docs repository50 , make a
change, and issue a pull request.
In order for the documentation team to accept your change, you must complete the MongoDB Contributor Agreement51 .
You can clone the repository by issuing the following command at your system shell:
git clone git://github.com/mongodb/docs.git

12.5.2 About the Documentation Process

The MongoDB Manual uses Sphinx52 , a sophisticated documentation engine built upon Python Docutils53 . The original reStructured Text54 files, as well as all necessary Sphinx extensions and build tools, are available in the same
repository as the documentation.
For more information on the MongoDB documentation process, see:
Style Guide and Documentation Conventions
This document provides an overview of the style for the MongoDB documentation stored in this repository. The
overarching goal of this style guide is to provide an accessible base style to ensure that our documentation is easy to
read, simple to use, and straightforward to maintain.
For information regarding the MongoDB Manual organization, see MongoDB Manual Organization (page 649).
44 http://tr.docs.mongodb.org
45 http://uk.docs.mongodb.org
46 http://groups.google.com/group/mongodb-translators
47 https://github.com/mongodb/docs
48 http://github.com/mongodb
49 https://github.com/
50 https://github.com/mongodb/docs
51 http://www.mongodb.com/contributor
52 http://sphinx-doc.org//
53 http://docutils.sourceforge.net/
54 http://docutils.sourceforge.net/rst.html

640

Chapter 12. About MongoDB Documentation

MongoDB Documentation, Release 2.4.9

Document History

2011-09-27: Document created with a (very) rough list of style guidelines, conventions, and questions.
2012-01-12: Document revised based on slight shifts in practice, and as part of an effort of making it easier for people
outside of the documentation team to contribute to documentation.
2012-03-21: Merged in content from the Jargon, and cleaned up style in light of recent experiences.
2012-08-10: Addition to the Referencing section.
2013-02-07: Migrated this document to the manual. Added map-reduce terminology convention. Other edits.
2013-11-15: Added new table of preferred terms.
Naming Conventions

This section contains guidelines on naming files, sections, documents and other document elements.
File naming Convention:
For Sphinx, all files should have a .txt extension.
Separate words in file names with hyphens (i.e. -.)
For most documents, file names should have a terse one or two word name that
scribes the material covered in the document.
Allow the path of the file within the
ument tree to add some of the required context/categorization.
For example its
ceptable
to
have
http://docs.mongodb.org/manualcore/sharding.rst
http://docs.mongodb.org/manualadministration/sharding.rst.

dedocacand

For tutorials, the full title of the document should be in the file name.
For example,
http://docs.mongodb.org/manualtutorial/replace-one-configuration-server-in-a-shard Phrase headlines and titles so users can determine what questions the text will answer, and material that will
be addressed, without needing them to read the content. This shortens the amount of time that people spend
looking for answers, and improvise search/scanning, and possibly SEO.
Prefer titles and headers in the form of Using foo over How to Foo.
When using target references (i.e. :ref: references in documents), use names that include enough context to
be intelligible through all documentation. For example, use replica-set-secondary-only-node as
opposed to secondary-only-node. This makes the source more usable and easier to maintain.
Style Guide

This includes the local typesetting, English, grammatical, conventions and preferences that all documents in the manual
should use. The goal here is to choose good standards, that are clear, and have a stylistic minimalism that does not
interfere with or distract from the content. A uniform style will improve user experience and minimize the effect of a
multi-authored document.
Punctuation
Use the Oxford comma.
Oxford commas are the commas in a list of things (e.g. something, something else, and another thing) before
the conjunction (e.g. and or or.).
Do not add two spaces after terminal punctuation, such as periods.

12.5. Contribute to the Documentation

641

MongoDB Documentation, Release 2.4.9

Place commas and periods inside quotation marks.

Headings Use title case for headings and document titles. Title case capitalizes the first letter of the first, last, and
all significant words.
Verbs Verb tense and mood preferences, with examples:
Avoid the first person. For example do not say, We will begin the backup process by locking the database, or
I begin the backup process by locking my database instance.
Use the second person. If you need to back up your database, start by locking the database first. In practice,
however, its more concise to imply second person using the imperative, as in Before initiating a backup, lock
the database.
When indicated, use the imperative mood. For example: Backup your databases often and To prevent data
loss, back up your databases.
The future perfect is also useful in some cases. For example, Creating disk snapshots without locking the
database will lead to an invalid state.
Avoid helper verbs, as possible, to increase clarity and concision. For example, attempt to avoid this does
foo and this will do foo when possible. Use does foo over will do foo in situations where this foos is
unacceptable.
Referencing
To refer to future or planned functionality in MongoDB or a driver, always link to the Jira case. The Manuals
conf.py provides an :issue: role that links directly to a Jira case (e.g. :issue:\SERVER-9001\).
For non-object references (i.e. functions, operators, methods, database commands, settings) always reference
only the first occurrence of the reference in a section. You should always reference objects, except in section
headings.
Structure references with the why first; the link second.
For example, instead of this:
Use the Convert a Replica Set to a Replicated Sharded Cluster (page 515) procedure if you have an existing
replica set.
Type this:
To deploy a sharded cluster for an existing replica set, see Convert a Replica Set to a Replicated Sharded Cluster
(page 515).
General Formulations
Contractions are acceptable insofar as they are necessary to increase readability and flow. Avoid otherwise.
Make lists grammatically correct.
Do not use a period after every item unless the list item completes the unfinished sentence before the list.
Use appropriate commas and conjunctions in the list items.
Typically begin a bulleted list with an introductory sentence or clause, with a colon or comma.
The following terms are one word:
standalone

642

Chapter 12. About MongoDB Documentation

MongoDB Documentation, Release 2.4.9

workflow
Use unavailable, offline, or unreachable to refer to a mongod instance that cannot be accessed. Do not
use the colloquialism down.
Always write out units (e.g. megabytes) rather than using abbreviations (e.g. MB.)
Structural Formulations
There should be at least two headings at every nesting level. Within an h2 block, there should be either: no
h3 blocks, 2 h3 blocks, or more than 2 h3 blocks.
Section headers are in title case (capitalize first, last, and all important words) and should effectively describe
the contents of the section. In a single document you should strive to have section titles that are not redundant
and grammatically consistent with each other.
Use paragraphs and paragraph breaks to increase clarity and flow. Avoid burying critical information in the
middle of long paragraphs. Err on the side of shorter paragraphs.
Prefer shorter sentences to longer sentences. Use complex formations only as a last resort, if at all (e.g. compound complex structures that require semi-colons).
Avoid paragraphs that consist of single sentences as they often represent a sentence that has unintentionally
become too complex or incomplete. However, sometimes such paragraphs are useful for emphasis, summary,
or introductions.
As a corollary, most sections should have multiple paragraphs.
For longer lists and more complex lists, use bulleted items rather than integrating them inline into a sentence.
Do not expect that the content of any example (inline or blocked) will be self explanatory. Even when it feels
redundant, make sure that the function and use of every example is clearly described.
ReStructured Text and Typesetting
Place spaces between nested parentheticals and elements in JavaScript examples. For example, prefer { [ a,
a, a ] } over {[a,a,a]}.
For underlines associated with headers in RST, use:
= for heading level 1 or h1s. Use underlines and overlines for document titles.
- for heading level 2 or h2s.
~ for heading level 3 or h3s.
for heading level 4 or h4s.
Use hyphens (-) to indicate items of an ordered list.
Place footnotes and other references, if you use them, at the end of a section rather than the end of a file.
Use the footnote format that includes automatic numbering and a target name for ease of use. For instance a
footnote tag may look like: [#note]_ with the corresponding directive holding the body of the footnote that
resembles the following: .. [#note].
Do not include ..

code-block::

[language] in footnotes.

As it makes sense, use the .. code-block:: [language] form to insert literal blocks into the text.
While the double colon, ::, is functional, the .. code-block:: [language] form makes the source
easier to read and understand.
For all mentions of referenced types (i.e. commands, operators, expressions, functions, statuses, etc.) use the
reference types to ensure uniform formatting and cross-referencing.
12.5. Contribute to the Documentation

643

MongoDB Documentation, Release 2.4.9

644

Chapter 12. About MongoDB Documentation

MongoDB Documentation, Release 2.4.9

Jargon and Common Terms

Preferred
Term
document

Concept

Dispreferred
Alternatives

Notes

A single, top-level object/record

in a MongoDB collection.

record, object,
row

Prefer document over object because of

concerns about cross-driver language handling
of objects. Reserve record for allocation of
storage. Avoid row, as possible.
Avoid genericizing database. Avoid using
database to refer to a server process or a data
set. This applies both to the datastoring
contexts as well as other (related) operational
contexts (command context,
authentication/authorization context.)
Avoid using instance, unless it modifies
something specifically. Having a descriptor for
a process/instance makes it possible to avoid
needing to make mongod or mongos plural.
Server and node are both vague and
contextually difficult to disambiguate with
regards to application servers, and underlying
hardware.
Avoid introducing unrelated terms for a single
field. In the documentation weve rarely had to
discuss the identifier of a field, so the extra
word here isnt burdensome.
Use to emphasize the difference between the
name of a field and its value For example,
_id is the field and the default value is an
ObjectId.

databaseA group of collections. Refers to

a group of data files. This is the
logical sense of the term
database.

instance

A daemon process. (e.g. mongos

or mongod)

process
(acceptable
sometimes), node
(never
acceptable),
server.

field
name

The identifier of a value in a

document.

key, column

field/value
The name/value pair that
describes a unit of data in
MongoDB.

key, slot, attribute

value
MongoDB

The data content of a field.

A group of processes, or
deployment that implement the
MongoDB interface.

data
mongo,
mongodb, cluster

subAn embedded or nested

documentdocument within a document or
an array.
mapAn operation performed by the
reduce mapReduce command.

embedded
document, nested
document
mapReduce, map
reduce,
map/reduce
grid, shard
cluster, set,
deployment

cluster

A sharded cluster.

sharded A sharded cluster.

cluster
replica A deployment of replicating
set
mongod programs that provide
redundancy and automatic
failover.
deA group of MongoDB processes,
ployor a standalone
12.5. Contribute
to themongod
Documentation
ment
instance.
data
The collection of physical
set
databases provided by a

Stylistic preference, mostly. In some cases its

useful to be able to refer generically to
instances (that may be either mongod or
mongos.)

Avoid confusion with the command, shell

helper, and driver interfaces. Makes it possible
to discuss the operation generally.
Cluster is a great word for a group of
processes; however, its important to avoid
letting the term become generic. Do not use for
any group of MongoDB processes or
deployments.

shard cluster,
cluster, sharded
system
set, replication
deployment

cluster, system

database, data

Typically in the form MongoDB deployment.

Includes standalones, replica sets and sharded645
clusters.
Important to keep the distinction between the
data provided by a mongod or a sharded cluster

MongoDB Documentation, Release 2.4.9

Database Systems and Processes

To indicate the entire database system, use MongoDB, not mongo or Mongo.
To indicate the database process or a server instance, use mongod or mongos. Refer to these as processes
or instances. Reserve database for referring to a database structure, i.e., the structure that holds collections
and refers to a group of files on disk.
Distributed System Terms
Refer to partitioned systems as sharded clusters. Do not use shard clusters or sharded systems.
Refer to configurations that run with replication as replica sets (or master/slave deployments) rather than
clusters or other variants.
Data Structure Terms
document refers to rows or records in a MongoDB database. Potential confusion with JSON Documents.
Do not refer to documents as objects, because drivers (and MongoDB) do not preserve the order of fields when
fetching data. If the order of objects matter, use an array.
field refers to a key or identifier of data within a MongoDB document.
value refers to the contents of a field.
sub-document describes a nested document.
Other Terms
Use example.net (and .org or .com if needed) for all examples and samples.
Hyphenate map-reduce in order to avoid ambiguous reference to the command name. Do not camel-case.
Notes on Specific Features

Geo-Location
1. While MongoDB is capable of storing coordinates in sub-documents, in practice, users should only store
coordinates in arrays. (See: DOCS-4155 .)
MongoDB Documentation Practices and Processes
This document provides an overview of the practices and processes.
Commits

When relevant, include a Jira case identifier in a commit message. Reference documentation cases when applicable,
but feel free to reference other cases from jira.mongodb.org56 .
Err on the side of creating a larger number of discrete commits rather than bundling large set of changes into one
commit.
55 https://jira.mongodb.org/browse/DOCS-41
56 http://jira.mongodb.org/

646

Chapter 12. About MongoDB Documentation

MongoDB Documentation, Release 2.4.9

For the sake of consistency, remove trailing whitespaces in the source file.
Hard wrap files to between 72 and 80 characters per-line.
Standards and Practices

At least two people should vet all non-trivial changes to the documentation before publication. One of the
reviewers should have significant technical experience with the material covered in the documentation.
All development and editorial work should transpire on GitHub branches or forks that editors can then merge
into the publication branches.
Collaboration

To propose a change to the documentation, do either of the following:

Open a ticket in the documentation project57 proposing the change. Someone on the documentation team will
make the change and be in contact with you so that you can review the change.
Using GitHub58 , fork the mongodb/docs repository59 , commit your changes, and issue a pull request. Someone
on the documentation team will review and incorporate your change into the documentation.
Builds

Building the documentation is useful because Sphinx60 and docutils can catch numerous errors in the format and
syntax of the documentation. Additionally, having access to an example documentation as it will appear to the users
is useful for providing more effective basis for the review process. Besides Sphinx, Pygments, and Python-Docutils,
the documentation repository contains all requirements for building the documentation resource.
Talk to someone on the documentation team if you are having problems running builds yourself.
Publication

The makefile for this repository contains targets that automate the publication process. Use make html to publish
a test build of the documentation in the build/ directory of your repository. Use make publish to build the full
contents of the manual from the current branch in the ../public-docs/ directory relative the docs repository.
Other targets include:
man - builds UNIX Manual pages for all Mongodb utilities.
push - builds and deploys the contents of the ../public-docs/.
pdfs - builds a PDF version of the manual (requires LaTeX dependencies.)
Branches

This section provides an overview of the git branches in the MongoDB documentation repository and their use.
57 https://jira.mongodb.org/browse/DOCS
58 https://github.com/
59 https://github.com/mongodb/docs
60 http://sphinx.pocoo.org/

12.5. Contribute to the Documentation

647

MongoDB Documentation, Release 2.4.9

At the present time, future work transpires in the master, with the main publication being current. As the
documentation stabilizes, the documentation team will begin to maintain branches of the documentation for specific
MongoDB releases.
Migration from Legacy Documentation

The MongoDB.org Wiki contains a wealth of information. As the transition to the Manual (i.e. this project and
resource) continues, its critical that no information disappears or goes missing. The following process outlines how
to migrate a wiki page to the manual:
1. Read the relevant sections of the Manual, and see what the new documentation has to offer on a specific topic.
In this process you should follow cross references and gain an understanding of both the underlying information
and how the parts of the new content relates its constituent parts.
2. Read the wiki page you wish to redirect, and take note of all of the factual assertions, examples presented by the
wiki page.
3. Test the factual assertions of the wiki page to the greatest extent possible. Ensure that example output is accurate.
In the case of commands and reference material, make sure that documented options are accurate.
4. Make corrections to the manual page or pages to reflect any missing pieces of information.
The target of the redirect need not contain every piece of information on the wiki page, if the manual as a
whole does, and relevant section(s) with the information from the wiki page are accessible from the target of the
redirection.
5. As necessary, get these changes reviewed by another writer and/or someone familiar with the area of the information in question.
At this point, update the relevant Jira case with the target that youve chosen for the redirect, and make the ticket
unassigned.
6. When someone has reviewed the changes and published those changes to Manual, you, or preferably someone
else on the team, should make a final pass at both pages with fresh eyes and then make the redirect.
Steps 1-5 should ensure that no information is lost in the migration, and that the final review in step 6 should be
trivial to complete.
Review Process

Types of Review The content in the Manual undergoes many types of review, including the following:
Initial Technical Review Review by an engineer familiar with MongoDB and the topic area of the documentation.
This review focuses on technical content, and correctness of the procedures and facts presented, but can improve any
aspect of the documentation that may still be lacking. When both the initial technical review and the content review
are complete, the piece may be published.
Content Review Textual review by another writer to ensure stylistic consistency with the rest of the manual. Depending on the content, this may precede or follow the initial technical review. When both the initial technical review
and the content review are complete, the piece may be published.

648

Chapter 12. About MongoDB Documentation

MongoDB Documentation, Release 2.4.9

Consistency Review This occurs post-publication and is content focused. The goals of consistency reviews are to
increase the internal consistency of the documentation as a whole. Insert relevant cross-references, update the style as
needed, and provide background fact-checking.
When possible, consistency reviews should be as systematic as possible and we should avoid encouraging stylistic and
information drift by editing only small sections at a time.
Subsequent Technical Review If the documentation needs to be updated following a change in functionality of the
server or following the resolution of a user issue, changes may be significant enough to warrant additional technical
review. These reviews follow the same form as the initial technical review, but is often less involved and covers a
smaller area.
Review Methods If youre not a usual contributor to the documentation and would like to review something, you
can submit reviews in any of the following methods:
If youre reviewing an open pull request in GitHub, the best way to comment is on the overview diff, which
you can find by clicking on the diff button in the upper left portion of the screen. You can also use the
following URL to reach this interface:
https://github.com/mongodb/docs/pull/[pull-request-id]/files

Replace [pull-request-id] with the identifier of the pull request. Make all comments inline, using
GitHubs comment system.
You may also provide comments directly on commits, or on the pull request itself but these commit-comments
are archived in less coherent ways and generate less useful emails, while comments on the pull request lead to
less specific changes to the document.
Leave feedback on Jira cases in the DOCS61 project. These are better for more general changes that arent
necessarily tied to a specific line, or affect multiple files.
Create a fork of the repository in your GitHub account, make any required changes and then create a pull request
with your changes.
If you insert lines that begin with any of the following annotations:
.. TODO:
TODO:
.. TODO
TODO

followed by your comments, it will be easier for the original writer to locate your comments. The two dots ..
format is a comment in reStructured Text, which will hide your comments from Sphinx and publication if youre
worried about that.
This format is often easier for reviewers with larger portions of content to review.
MongoDB Manual Organization
This document provides an overview of the global organization of the documentation resource. Refer to the notes
below if you are having trouble understanding the reasoning behind a files current location, or if you want to add new
documentation but arent sure how to integrate it into the existing resource.
If you have questions, dont hesitate to open a ticket in the Documentation Jira Project62 or contact the documentation
team63 .
61 http://jira.mongodb.org/browse/DOCS
62 https://jira.mongodb.org/browse/DOCS
63 [email protected]

12.5. Contribute to the Documentation

649

MongoDB Documentation, Release 2.4.9

Global Organization

Indexes
and
Experience The
documentation
project
has
two
index
files:
http://docs.mongodb.org/manualcontents.txt and http://docs.mongodb.org/manualindex.txt.
The contents file provides the documentations tree structure, which Sphinx uses to create the left-pane navigational
structure, to power the Next and Previous page functionality, and to provide all overarching outlines of the
resource. The index file is not included in the contents file (and thus builds will produce a warning here) and is
the page that users first land on when visiting the resource.
Having separate contents and index files provides a bit more flexibility with the organization of the resource while
also making it possible to customize the primary user experience.
Topical Organization The placement of files in the repository depends on the type of documentation rather than the
topic of the content. Like the difference between contents.txt and index.txt, by decoupling the organization
of the files from the organization of the information the documentation can be more flexible and can more adequately
address changes in the product and in users needs.
Files in the source/ directory represent the tip of a logical tree of documents, while directories are containers of
types of content. The administration and applications directories, however, are legacy artifacts and with a
few exceptions contain sub-navigation pages.
With several exceptions in the reference/ directory, there is only one level of sub-directories in the source/
directory.
Tools

The organization of the site, like all Sphinx sites derives from the toctree64 structure. However, in order to annotate
the table of contents and provide additional flexibility, the MongoDB documentation generates toctree65 structures
using data from YAML files stored in the source/includes/ directory. These files start with ref-toc or toc
and generate output in the source/includes/toc/ directory. Briefly this system has the following behavior:
files that start with ref-toc refer to the documentation of API objects (i.e. commands, operators and methods),
and the build system generates files that hold toctree66 directives as well as files that hold tables that list
objects and a brief description.
files that start with toc refer to all other documentation and the build system generates files that hold
toctree67 directives as well as files that hold definition lists that contain links to the documents and short
descriptions the content.
file names that have spec following toc or ref-toc will generate aggregated tables or definition lists and
allow ad-hoc combinations of documents for landing pages and quick reference guides.
MongoDB Documentation Build System
This document contains more direct instructions for building the MongoDB documentation.
Getting Started

Install Dependencies The MongoDB Documentation project depends on the following tools:
64 http://sphinx-doc.org/markup/toctree.html#directive-toctree
65 http://sphinx-doc.org/markup/toctree.html#directive-toctree
66 http://sphinx-doc.org/markup/toctree.html#directive-toctree
67 http://sphinx-doc.org/markup/toctree.html#directive-toctree

650

Chapter 12. About MongoDB Documentation

MongoDB Documentation, Release 2.4.9

GNU Make
GNU Tar
Python
Git
Sphinx (documentation management toolchain)
Pygments (syntax highlighting)
PyYAML (for the generated tables)
Droopy (Python package for static text analysis)
Fabric (Python package for scripting and orchestration)
Inkscape (Image generation.)
python-argparse (For Python 2.6.)
LaTeX/PDF LaTeX (typically texlive; for building PDFs)
Common Utilities (rsync, tar, gzip, sed)
OS X Install Sphinx, Docutils, and their dependencies with easy_install the following command:
easy_install Sphinx Jinja2 Pygments docutils

PyYAML droopy fabric

Feel free to use pip rather than easy_install to install python packages.
To generate the images used in the documentation, download and install Inkscape68 .
Optional
To generate PDFs for the full production build, install a TeX distribution (for building the PDF.) If you do not have a
LaTeX installation, use MacTeX69 . This is only required to build PDFs.

Arch Linux Install packages from the system repositories with the following command:
pacman -S python2-sphinx python2-yaml inkscape python2-pip

Then install the following Python packages:

pip install droopy fabric

Optional
To generate PDFs for the full production build, install the following packages from the system repository:
pacman -S texlive-bin texlive-core texlive-latexextra

Debian/Ubuntu Install the required system packages with the following command:
apt-get install python-sphinx python-yaml python-argparse inkscape python-pip

Then install the following Python packages:

68 http://inkscape.org/download/
69 http://www.tug.org/mactex/2011/

12.5. Contribute to the Documentation

651

MongoDB Documentation, Release 2.4.9

pip install droopy fabric

Optional
To generate PDFs for the full production build, install the following packages from the system repository:
apt-get install texlive-latex-recommended texlive-latex-recommended

Setup and Configuration Clone the repository:

git clone git://github.com/mongodb/docs.git

Then run the bootstrap.py script in the docs/ repository, to configure the build dependencies:
python bootstrap.py

This downloads and configures the mongodb/docs-tools70 repository, which contains the authoritative build system
shared between branches of the MongoDB Manual and other MongoDB documentation projects.
You can run bootstrap.py regularly to update build system.
Building the Documentation

The MongoDB documentation build system is entirely accessible via make targets. For example, to build an HTML
version of the documentation issue the following command:
make html

You can find the build output in build/<branch>/html, where <branch> is the name of the current branch.
In addition to the html target, the build system provides the following targets:
publish Builds and integrates all output for the production build.
Build output is in
build/public/<branch>/. When you run publish in the master, the build will generate
some output in build/public/.
push; stage Uploads the production build to the production or staging web servers. Depends on publish. Requires access production or staging environment.
push-all; stage-all Uploads the entire content of build/public/ to the web servers.
publish. Not used in common practice.

Depends on

push-with-delete; stage-with-delete Modifies the action of push and stage to remove remote file
that dont exist in the local build. Use with caution.
html; latex; dirhtml; epub; texinfo; man; json These are standard targets derived from the default
Sphinx Makefile, with adjusted dependencies. Additionally, for all of these targets you can append -nitpick
to increase Sphinxs verbosity, or -clean to remove all Sphinx build artifacts.
latex performs several additional post-processing steps on .tex output generated by Sphinx. This target will
also compile PDFs using pdflatex.
html and man also generates a .tar.gz file of the build outputs for inclusion in the final releases.
70 http://github.com/mongodb/docs-tools/

652

Chapter 12. About MongoDB Documentation

MongoDB Documentation, Release 2.4.9

Build Mechanics and Tools

Internally the build system has a number of components and processes. See the docs-tools README71 for more
information on the internals. This section documents a few of these components from a very high level and lists useful
operations for contributors to the documentation.
Fabric Fabric is an orchestration and scripting package for Python. The documentation uses Fabric to handle the
deployment of the build products to the web servers and also unifies a number of independent build operations. Fabric
commands have the following form:
fab <module>.<task>[:<argument>]

The <argument> is optional in most cases. Additionally some tasks are available at the root level, without a module.
To see a full list of fabric tasks, use the following command:
fab -l

You can chain fabric tasks on a single command line, although this doesnt always make sense.
Important fabric tasks include:
tools.bootstrap Runs the bootstrap.py script. Useful for re-initializing the repository without needing to
be in root of the repository.
tools.dev; tools.reset tools.dev switches the origin remote of the docs-tools checkout in build
directory, to ../docs-tools to facilitate build system testing and development. tools.reset resets the
origin remote for normal operation.
tools.conf tools.conf returns the content of the configuration object for the current project. These data are
useful during development.
stats.report:<filename> Returns, a collection of readability statistics.
source/ tree.

Specify file names relative to

make Provides a thin wrapper around Make calls. Allows you to start make builds from different locations in the
project repository.
process.refresh_dependencies Updates the time stamp of .txt source files with changed include files, to
facilitate Sphinxs incremental rebuild process. This task runs internally as part of the build process.
Buildcloth Buildcloth72 is a meta-build tool, used to generate Makefiles programmatically. This makes the build
system easier to maintain, and makes it easier to use the same fundamental code to generate various branches of the
Manual as well as related documentation projects. See makecloth/ in the docs-tools repository73 for the relevant code.
Running make with no arguments will regenerate these parts of the build system automatically.
Rstcloth Rstcloth74 is a library for generating reStructuredText programmatically. This makes it possible to generate
content for the documentation, such as tables, tables of contents, and API reference material programmatically and
transparently. See rstcloth/ in the docs-tools repository75 for the relevant code.
If you have any questions, please feel free to open a Jira Case76 .
71 https://github.com/mongodb/docs-tools/blob/master/README.rst
72 https://pypi.python.org/pypi/buildcloth/
73 https://github.com/mongodb/docs-tools/tree/master/makecloth
74 https://pypi.python.org/pypi/rstcloth
75 https://github.com/mongodb/docs-tools/tree/master/rstcloth
76 https://jira.mongodb.org/browse/DOCS

12.5. Contribute to the Documentation

653

MongoDB Documentation, Release 2.4.9

654

Chapter 12. About MongoDB Documentation

Index

Symbols

_id, 320
_id index, 320
<database>.system.indexes (MongoDB reporting output),
227
<database>.system.js (MongoDB reporting output), 227
<database>.system.namespaces (MongoDB reporting
output), 227
<database>.system.profile (MongoDB reporting output),
227
<database>.system.users (MongoDB reporting output),
272
<database>.system.users.pwd (MongoDB reporting output), 273
<database>.system.users.roles (MongoDB reporting output), 273
<database>.system.users.user (MongoDB reporting output), 273
<database>.system.users.userSource (MongoDB reporting output), 273
0 (error code), 235
100 (error code), 236
12 (error code), 236
14 (error code), 236
2 (error code), 235
20 (error code), 236
3 (error code), 236
4 (error code), 236
45 (error code), 236
47 (error code), 236
48 (error code), 236
49 (error code), 236
5 (error code), 236

balancing, 500
configure, 529
internals, 501
migration, 502
operations, 531
secondary throttle, 530

chunks._id (MongoDB reporting output), 127

chunks.data (MongoDB reporting output), 128
chunks.files_id (MongoDB reporting output), 128
chunks.n (MongoDB reporting output), 128
clusterAdmin (user role), 269
collection
system, 227
compound index, 322
config, 488
config (MongoDB reporting output), 548
config databases, 488
config servers, 487
config.changelog (MongoDB reporting output), 548
config.changelog._id (MongoDB reporting output), 549
config.changelog.clientAddr (MongoDB reporting output), 549
config.changelog.details (MongoDB reporting output),
549
config.changelog.ns (MongoDB reporting output), 549
config.changelog.server (MongoDB reporting output),
549
config.changelog.time (MongoDB reporting output), 549
config.changelog.what (MongoDB reporting output), 549
config.chunks (MongoDB reporting output), 549
config.collections (MongoDB reporting output), 550
A
config.databases (MongoDB reporting output), 550
admin.system.users.otherDBRoles (MongoDB reporting config.lockpings (MongoDB reporting output), 550
output), 274
config.locks (MongoDB reporting output), 550
administration tutorials, 187
config.mongos (MongoDB reporting output), 551
ARBITER (replica set state), 474
config.settings (MongoDB reporting output), 551
config.shards (MongoDB reporting output), 552

655

MongoDB Documentation, Release 2.4.9

config.tags (MongoDB reporting output), 552

config.version (MongoDB reporting output), 552
connection pooling
read operations, 38
consistency
rollbacks, 401
crud
write operations, 43

D
data_binary (BSON type), 228
data_date (BSON type), 228
data_maxkey (BSON type), 230
data_minkey (BSON type), 230
data_oid (BSON type), 229
data_ref (BSON type), 229
data_regex (BSON type), 229
data_timestamp (BSON type), 228
data_undefined (BSON type), 230
database, 488
local, 472
database references, 124
dbAdmin (user role), 269
dbAdminAnyDatabase (user role), 271
DBRef, 124
development tutorials, 189
DOWN (replica set state), 475

E
EDITOR, 569
environment variable
EDITOR, 569
HOME, 212

F
failover
replica set, 397
FATAL (replica set state), 475
files._id (MongoDB reporting output), 128
files.aliases (MongoDB reporting output), 128
files.chunkSize (MongoDB reporting output), 128
files.contentType (MongoDB reporting output), 128
files.filename (MongoDB reporting output), 128
files.length (MongoDB reporting output), 128
files.md5 (MongoDB reporting output), 128
files.metadata (MongoDB reporting output), 128
files.uploadDate (MongoDB reporting output), 128
fundamentals
sharding, 489

G
geospatial queries, 355
exact, 355

656

GridFS, 104, 127

chunks collection, 127
collections, 127
files collection, 128
index, 105
initialize, 104

H
HOME, 212

I
index
_id, 320
background creation, 336
compound, 322, 341
create, 339, 341
create in background, 345
drop duplicates, 338, 342
duplicates, 338, 342
embedded fields, 321
hashed, 333, 343
list indexes, 348
measure use, 349
monitor index building, 348
multikey, 324
name, 338
options, 336
overview, 313
rebuild, 347
remove, 347
replica set, 343
sort order, 323
sparse, 335, 342
subdocuments, 321
TTL index, 334
unique, 334, 341
index types, 319
primary key, 320
installation, 3
installation guides, 3
installation tutorials, 3
internals
config database, 547

L
load balancer, 490
local database, 472
local.oplog.$main (MongoDB reporting output), 474
local.oplog.rs (MongoDB reporting output), 473
local.replset.minvalid (MongoDB reporting output), 473
local.slaves (MongoDB reporting output), 473, 474
local.sources (MongoDB reporting output), 474
local.startup_log (MongoDB reporting output), 472
local.startup_log._id (MongoDB reporting output), 473
Index

MongoDB Documentation, Release 2.4.9

local.startup_log.buildinfo (MongoDB reporting output),

473
local.startup_log.cmdLine (MongoDB reporting output),
473
local.startup_log.hostname (MongoDB reporting output),
473
local.startup_log.pid (MongoDB reporting output), 473
local.startup_log.startTime (MongoDB reporting output),
473
local.startup_log.startTimeLocal (MongoDB reporting
output), 473
local.system.replset (MongoDB reporting output), 473
local.system.replset._id (MongoDB reporting output),
467
local.system.replset.members (MongoDB reporting output), 467
local.system.replset.members[n]._id (MongoDB reporting output), 467
local.system.replset.members[n].arbiterOnly (MongoDB
reporting output), 468
local.system.replset.members[n].buildIndexes
(MongoDB reporting output), 468
local.system.replset.members[n].hidden (MongoDB reporting output), 468
local.system.replset.members[n].host (MongoDB reporting output), 468
local.system.replset.members[n].priority (MongoDB reporting output), 469
local.system.replset.members[n].slaveDelay (MongoDB
reporting output), 469
local.system.replset.members[n].tags (MongoDB reporting output), 469
local.system.replset.members[n].votes
(MongoDB
reporting output), 469
local.system.replset.settings (MongoDB reporting output), 470
local.system.replset.settings.chainingAllowed
(MongoDB reporting output), 470
local.system.replset.settings.getLastErrorDefaults (MongoDB reporting output), 470
local.system.replset.settings.getLastErrorModes (MongoDB reporting output), 470

M
mongos, 490, 496
mongos load balancer, 490

N
namespace
local, 472
system, 227
nearest (read preference mode), 477

P
primary (read preference mode), 476
PRIMARY (replica set state), 474
primaryPreferred (read preference mode), 476

Q
query optimizer, 37

R
read (user role), 268
read operation
architecture, 38
connection pooling, 38
read operations
query, 31
read preference, 406
background, 406
behavior, 408
member selection, 408
modes, 476
mongos, 410
nearest, 408
ping time, 408
semantics, 476
sharding, 410
tag sets, 408, 450
readAnyDatabase (user role), 271
readWrite (user role), 268
readWriteAnyDatabase (user role), 271
RECOVERING (replica set state), 475
references, 124
replica set
elections, 397
failover, 397
index, 343
local database, 472
network partitions, 397
reconfiguration, 454
resync, 449, 450
rollbacks, 401
security, 240
sync, 410, 449
tag sets, 450
replica set members
arbiters, 388
delayed, 387
hidden, 387
non-voting, 400
ROLLBACK (replica set state), 475
rollbacks, 401

S
secondary (read preference mode), 476

Index

657

MongoDB Documentation, Release 2.4.9

SECONDARY (replica set state), 474

secondary throttle, 530
secondaryPreferred (read preference mode), 477
security
replica set, 240
shard key, 492
cardinality, 512
query isolation, 494
write scaling, 493
sharded clusters, 506
sharding
chunk size, 504
config database, 547
config servers, 487
localhost, 242
shard key, 492
shard key indexes, 505
shards, 485
shards, 485
SHUNNED (replica set state), 475
slaveOk, 406
STARTUP (replica set state), 475
STARTUP2 (replica set state), 475
system
collections, 227
namespace, 227
system.profile.client (MongoDB reporting output), 233
system.profile.command (MongoDB reporting output),
232
system.profile.cursorid (MongoDB reporting output), 232
system.profile.keyUpdates (MongoDB reporting output),
233
system.profile.lockStats (MongoDB reporting output),
233
system.profile.lockStats.timeAcquiringMicros
(MongoDB reporting output), 233
system.profile.lockStats.timeLockedMicros (MongoDB
reporting output), 233
system.profile.millis (MongoDB reporting output), 233
system.profile.moved (MongoDB reporting output), 232
system.profile.nmoved (MongoDB reporting output), 232
system.profile.nreturned (MongoDB reporting output),
233
system.profile.ns (MongoDB reporting output), 232
system.profile.nscanned (MongoDB reporting output),
232
system.profile.ntoreturn (MongoDB reporting output),
232
system.profile.ntoskip (MongoDB reporting output), 232
system.profile.numYield (MongoDB reporting output),
233
system.profile.nupdated (MongoDB reporting output),
232
system.profile.op (MongoDB reporting output), 231

658

system.profile.query (MongoDB reporting output), 232

system.profile.responseLength (MongoDB reporting output), 233
system.profile.scanAndOrder (MongoDB reporting output), 232
system.profile.ts (MongoDB reporting output), 231
system.profile.updateobj (MongoDB reporting output),
232
system.profile.user (MongoDB reporting output), 234

T
tag sets, 408
configuration, 450
text search tutorials, 189
TTL index, 334
tutorials, 186
administration, 187
development patterns, 189
installation, 3
text search, 189

U
UNKNOWN (replica set state), 475
userAdmin (user role), 269
userAdminAnyDatabase (user role), 271

W
write concern, 46
write operations, 43

Index