Terraform Resources

Last Updated : 23 Dec, 2024

A Terraform resource is like a building block in a blueprint for your infrastructure. Think of it as a specific piece of the infrastructure that you want to create, manage, or update. For example, it could be a virtual machine, a database, a storage bucket or a load balancer.

When using Terraform, you write down what resources you need in a configuration file. For each resource, you describe its details such as the type of machine, how much storage it needs or its location in the cloud. Terraform then creates or updates those resources to match your description.

Here’s a simpler analogy: If building infrastructure is like constructing a house, then a Terraform resource is like an individual room (kitchen, bedroom, bathroom). You define what each room should look like and Terraform helps you "build" those rooms in the real world (your cloud or data center).

Key Points

What you can define: Anything in your infrastructure servers, databases, networks, etc.
Customization: You decide the specifics, like size, location, or configuration.
Automation: Terraform handles creating, updating, or deleting them for you.

For example, in Terraform, if you want a virtual machine, you’d define it like this:

resource "aws_instance" "my_vm" {  ami = "ami-12345678" # The operating system image  instance_type = "t2.micro"     # Size of the virtual machine}

This tells Terraform: "I want a small virtual machine using this operating system image." Terraform then makes it happen automatically.

Syntax of Terraform Resource Block

A resource block in Terraform is used to define individual pieces of infrastructure.To create an AWS EC2 instance, you would use a resource block like this:

resource "aws_instance" "example" {
ami = "ami-0c55b159cbfafe1f0"
instance_type = "t2.micro"
}

Here’s how it works:

Resource keyword:This indicates that you are defining a resource.
Type and Name:
- aws_instance is the type of resource you are creating—in this case, an EC2 instance.
- "example" is a unique name you assign to this resource so you can reference it later in the configuration.
Arguments:
- ami: This specifies the Amazon Machine Image (AMI) ID, which is the template for the instance.
- instance_type: This defines the size or type of the instance, like t2.micro for a small instance.

Resource blocks always start with the resource keyword, followed by the type and name, and then the arguments inside curly braces {}. Some blocks may also include additional nested configurations like network_interface for networking details.

Once you’ve written the resource block, you can apply the configuration to create the resource by running the following commands:

terraform init
terraform apply

These steps initialize Terraform and provision the EC2 instance using the settings in the resource block.

Use of Terraform Resources Documentation

Using Terraform’s resource documentation is a straightforward way to learn how to define and work with specific resources in your infrastructure setup. Here’s how you can use it effectively:

1. Open the terraform documentation

Go to the official Terraform documentation site at https://registry.terraform.io/. This is where you’ll find details about all Terraform providers and their resources.

2. Search for the resource you need

Use the search bar to find the resource you’re interested in, such as aws_instance for creating an EC2 instance on AWS.
Click on the provider (e.g., AWS, Azure, or Google Cloud) to see a list of all supported resources.

3. Understand the key sections of the documentation

Each resource page is divided into important sections:

Resource Name: This tells you the name you’ll use to define the resource in your Terraform code, like aws_instance.
Example Usage: The documentation provides examples to help you get started quickly. For instance, to create an EC2 instance:

resource "aws_instance" "example" {
  ami           = "ami-12345678"
  instance_type = "t2.micro"
}

Arguments Reference: This section lists the options you can set for the resource, including:
- Required arguments: These must be included (e.g., ami and instance_type for aws_instance).
- Optional arguments: These are additional settings you can customize if needed.
Attributes Reference: These are the values Terraform generates after creating the resource, such as the public IP address (public_ip) or resource ID (id).
Nested Blocks: Some resources include blocks for more detailed configurations, like network_interface for EC2 instances.
Import Instructions: If you have an existing resource outside Terraform, this section explains how to import it into Terraform.

4. Identify Required and Optional settings

Read the documentation carefully to understand which settings are mandatory and which are optional. For example, to create an EC2 instance, you must provide the ami ID and instance_type. Optional arguments, like tags, can be added for customization.

5. Start with the Examples

Use the examples provided in the documentation as a base for your configuration. Modify them to match your requirements. For example:

resource "aws_instance" "my_instance" {
  ami           = "ami-12345678"
  instance_type = "t2.micro"
  tags = {
    Name = "MyInstance"
  }
}

6. Use output attributes in your code

Once a resource is created, you can use its attributes in other parts of your configuration. For example, to output the public IP of an EC2 instance:

output "instance_ip" {
  value = aws_instance.my_instance.public_ip
}

7. Explore advanced configurations

Many resource pages include advanced examples, like setting up multiple volumes, configuring networking, or using lifecycle rules. These can be helpful for complex setups.

8. Stay updated

Terraform providers are updated regularly. Check the documentation often to ensure you’re using the latest features and arguments for your resources.

Resource Types and Arguments in Terraform

1. Resource Types

A resource type represents a specific infrastructure component, like a virtual machine, storage bucket, or database, that Terraform manages. Each resource type is tied to a provider, such as AWS, Azure, or Google Cloud.

Examples of Resource Types

aws_instance: Represents an Amazon EC2 instance.
google_storage_bucket: Defines a storage bucket in Google Cloud.
azurerm_virtual_network: Refers to a virtual network in Azure.

When using a resource type, you define its properties and configurations based on the provider’s requirements.

2. Arguments

Arguments are the settings or properties that define the characteristics of a resource. They are specified within the resource block.

1. Required Arguments: These are mandatory and must be provided. For example, to create an EC2 instance (aws_instance), you need to specify the ami (Amazon Machine Image ID) and instance_type:

resource "aws_instance" "my_instance" {
  ami           = "ami-0c55b159cbfafe1f0"
  instance_type = "t2.micro"
}

2. Optional Arguments: These allow for additional customization but are not required. For example, adding tags to an EC2 instance is optional:

resource "aws_instance" "my_instance" {
  ami           = "ami-0c55b159cbfafe1f0"
  instance_type = "t2.micro"
  tags = {
    Name = "WebServer"
  }
}

3. Computed Arguments:These are automatically generated by the provider and can’t be directly set. For instance, the public_ip of an EC2 instance is a computed value.

Using Output Attributes

After creating a resource, Terraform provides details about it through output attributes. These can be used in other parts of your configuration. For example:

public_ip for an EC2 instance
bucket_domain_name for an S3 bucket

Resource Dependencies in Terraform

In Terraform, resource dependencies help define the order in which resources should be created, updated, or destroyed. These dependencies ensure that resources are provisioned in the correct sequence and prevent issues where one resource depends on another.

Types of Dependencies

1. Implicit Dependencies:Terraform automatically detects dependencies when one resource references another. For instance, if an EC2 instance depends on a security group, Terraform knows to create the security group first.

resource "aws_instance" "example" {
  ami = "ami-0c55b159cbfafe1f0"
  vpc_security_group_ids = [aws_security_group.example.id]
}

resource "aws_security_group" "example" { ... }

2. Explicit Dependencies:In some cases, Terraform may not automatically detect a dependency, so you can explicitly declare it using the depends_on argument.

resource "aws_s3_bucket" "example" { ... }

resource "aws_s3_bucket_policy" "example" {
  depends_on = [aws_s3_bucket.example]
}

Meta Arguments in Terraform

Meta-arguments in Terraform are special settings that control how Terraform behaves when managing infrastructure. These arguments aren't directly part of the resource or data blocks but help in configuring resources more efficiently, especially in complex scenarios.

Key Meta Arguments

1. depends_on: This argument defines explicit dependencies between resources. It ensures that one resource is created or destroyed before or after another, even if there's no direct reference between them.

resource "aws_s3_bucket" "example" { ... }

resource "aws_s3_bucket_policy" "example" {
  depends_on = [aws_s3_bucket.example]
}

This ensures the bucket is created before applying the policy.

2. count: The count argument allows you to create multiple instances of a resource. The value of count specifies how many instances Terraform should create.

resource "aws_instance" "example" {
  count         = 3
  ami           = "ami-0c55b159cbfafe1f0"
  instance_type = "t2.micro"
}

3. for_each: Similar to count, but more flexible. With for_each, you can create resources based on a map or a list, which allows you to control each instance more precisely.

resource "aws_security_group" "example" {
  for_each = toset(["web", "db", "app"])
  name     = "${each.value}-security-group"
}

This example creates a security group for each item in the set.

4. provider: The provider argument allows you to specify which provider to use for a resource. This is useful when working with multiple providers in the same configuration.

resource "aws_instance" "example" {
  provider = aws.us_east_1
  ami      = "ami-0c55b159cbfafe1f0"
  instance_type = "t2.micro"
}

This tells Terraform to use the AWS provider in the us_east_1 region for this resource.

5. lifecycle: The lifecycle argument controls how Terraform handles the creation, modification, and destruction of a resource. Some useful settings include:

create_before_destroy: Ensures that a new resource is created before the old one is destroyed.
prevent_destroy: Prevents a resource from being destroyed.
ignore_changes: Tells Terraform to ignore specific changes to a resource.

Example:

resource "aws_instance" "example" {
  ami           = "ami-0c55b159cbfafe1f0"
  instance_type = "t2.micro"

  lifecycle {
    prevent_destroy = true
  }
}

6. metadata: The metadata argument is often used for tagging or providing additional information about resources.

resource "aws_instance" "example" {
  ami           = "ami-0c55b159cbfafe1f0"
  instance_type = "t2.micro"
  tags = {
    "Name" = "WebServer"
  }
}

Local-Only Resources in Terraform

In Terraform, local-only resources are defined using the null_resource type. These resources don’t represent actual infrastructure, but are useful for managing actions that stay within Terraform’s state, without affecting any external systems.

A common use for null_resource is when you want to run an action or trigger something every time you apply your Terraform configuration. For instance, you can set up a trigger that ensures the resource is recreated on every terraform apply, even if the rest of the configuration hasn't changed.

Here’s an example of a null_resource:

resource "null_resource" "example" {
  triggers = {
    # This will ensure the resource is recreated each time terraform apply is run
    always_run = "${timestamp()}"
  }
}

In this case, the always_run trigger is set to the current timestamp. This guarantees that the null_resource will be recreated on every apply, allowing you to run certain tasks or actions each time, even when no changes are made to your actual infrastructure.

Create Terraform Resources

Install Terraform: Start by downloading and installing Terraform from its official website.
Create a Configuration File: Set up a directory for your project and create a .tf file, such as main.tf.
Set Up the Provider: Define the cloud provider you’re using, like AWS. Here’s an example:

provider "aws" {
  region = "us-west-2"
}

Define the Resources: Add the resources you want to create, such as an EC2 instance:

resource "aws_instance" "example" {
  ami           = "ami-0c55b159cbfafe1f0"
  instance_type = "t2.micro"
}

Initialize the Project: Run terraform init to set up your project and download the necessary plugins.
Preview the Changes: Use terraform plan to see what Terraform will create before applying any changes.
Apply the Configuration: Run terraform apply to create the resources as defined in your configuration.
Remove Resources (if needed): If you want to delete the resources, use terraform destroy.

Custom Condition Checks for Terraform Resources

In Terraform, you can use various techniques to manage resources based on specific conditions. Here's a simplified overview:

1. Using count for Conditional Resources:The count argument allows you to create resources based on a condition. For example, if you want to create an instance only when a variable is set to true, you can use:

resource "aws_instance" "example" {
  count = var.create_instance ? 1 : 0
  ami = "ami-0c55b159cbfafe1f0"
  instance_type = "t2.micro"
}

2. Using for_each for Conditional Resource Creation: With for_each, you can create multiple resources only if a certain condition is met. For instance:

resource "aws_instance" "example" {
  for_each = var.instances_to_create != 0 ? { for i in range(var.instances_to_create) : i => i } : {}
  ami = "ami-0c55b159cbfafe1f0"
  instance_type = "t2.micro"
}

3. Ignoring Changes with lifecycle: The lifecycle block allows you to ignore changes to certain resource attributes after they’ve been created. For example:

resource "aws_instance" "example" {
  ami = "ami-0c55b159cbfafe1f0"
  instance_type = "t2.micro"
  lifecycle {
    ignore_changes = [ami]
  }
}

4. Controlling Creation Order with depends_on: If you need to control the order in which resources are created, use depends_on to ensure one resource is created before another:

resource "aws_instance" "example" {
  depends_on = [aws_s3_bucket.example]
  ami = "ami-0c55b159cbfafe1f0"
  instance_type = "t2.micro"
}

5. Using locals for Dynamic Conditions: You can use locals to define conditions dynamically based on variables. For example, create a resource only if the environment is set to "production":

locals {
  should_create = var.environment == "production" ? true : false
}
resource "aws_instance" "example" {
  count = local.should_create ? 1 : 0
  ami = "ami-0c55b159cbfafe1f0"
  instance_type = "t2.micro"
}

Terraform Resource Operation Timeouts

In Terraform, you can control how long Terraform waits for a resource operation to complete, such as creating, updating, or deleting a resource, by using the timeouts argument. This is particularly helpful when a resource might take longer than expected to be provisioned or modified.

Example:

resource "aws_instance" "example" {  ami = "ami-0c55b159cbfafe1f0"  instance_type = "t2.micro"  timeouts {  create = "30m"    update = "20m"    delete = "10m"  }}

In this example:

create: Terraform will wait for up to 30 minutes to create the EC2 instance.
update: Terraform will wait for up to 20 minutes for an update.
delete: Terraform will wait for up to 10 minutes to delete the resource.

If the resource operation takes longer than the specified time, Terraform will stop and report an error. If you don't set timeouts, Terraform will use default waiting times. This feature helps avoid unnecessary failures for resources that need more time to be processed.