Kubernetes Summary: Concepts and

Configurations

Contents
1 Introduction 1

2 Kubernetes Cluster Overview 2

2.1 Components of a Kubernetes Cluster . . . . . . . . . . . . . . . . . . . 2
2.2 Real-World Analogy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

3 Kubernetes vs. Docker 2

4 Minikube Overview 3
4.1 Minikube with Docker . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
4.2 Minikube Setup Steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

5 Kubernetes Configuration 3
5.1 Deployments for Stateless Apps . . . . . . . . . . . . . . . . . . . . . . . 4
5.2 StatefulSets for Stateful Apps . . . . . . . . . . . . . . . . . . . . . . . . 4
5.3 Services for Exposing Pods . . . . . . . . . . . . . . . . . . . . . . . . . . 5
5.4 ConfigMaps and Secrets . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

6 Demo Project Structure 6

6.1 Project Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
6.2 Key Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

7 Troubleshooting Example 6

8 Conclusion 6

1 Introduction
This document summarizes the key theoretical concepts and configuration de-
tails about Kubernetes, as discussed in a conversation about setting up and un-
derstanding a Kubernetes cluster. It covers the structure of a Kubernetes cluster,
the role of its components, and practical configurations for deploying applica-
tions, with a focus on a demo project involving MongoDB and a web application.

1
2 Kubernetes Cluster Overview
A Kubernetes cluster is a group of computers (nodes) that work together to run
containerized applications, ensuring scalability, high availability, and efficient
resource management.

2.1 Components of a Kubernetes Cluster

• Master Node (Control Plane): The ”boss” that manages the cluster, han-
dling tasks like scheduling, scaling, and monitoring. It includes:

– API Server: Acts as the communication hub, accepting commands

from kubectl to manage the cluster.
– etcd: A key-value store that holds the cluster’s state (e.g., desired vs.
actual state of pods).
– Scheduler: Assigns pods to worker nodes based on resource availabil-
ity.
– Controller Manager: Ensures the desired state is maintained (e.g.,
creating new pods if one fails).

• Worker Nodes: The ”doers” that run the actual applications in containers.
Each worker node has:

– Kubelet: A process that communicates with the master node and man-
ages containers on the node.
– Container Runtime (e.g., Docker, containerd): Runs and manages
containers.
– Pods: The smallest deployable units, each containing one or more con-
tainers.

2.2 Real-World Analogy

Think of a Kubernetes cluster as a restaurant:
• Master Node: The head chef who plans the menu and assigns tasks.

• Worker Nodes: The cooks who prepare the food.

• Kubelet: The recipe book each cook follows, ensuring tasks are done cor-
rectly.

• Pods/Containers: The dishes being prepared, served to customers (users).

3 Kubernetes vs. Docker

• Docker: A tool for creating, running, and managing containers and images.
It lacks built-in features for high availability or scalability.

• Kubernetes: An orchestration platform that manages containers across

multiple nodes, providing:

2
 – Scalability: Automatically adds/removes containers based on demand.
– High Availability: Restarts failed containers and ensures redundancy.
– Load Balancing: Distributes traffic evenly across containers.
– Distribution: Spreads containers across nodes to optimize resources.

4 Minikube Overview
Minikube is a tool for running a single-node Kubernetes cluster locally, ideal for
learning and testing. It can run inside a Docker container or a virtual machine
(e.g., VirtualBox).

4.1 Minikube with Docker

• Minikube uses Docker as the container runtime to run application contain-
ers.

• Alternatively, Minikube itself can run as a Docker container, hosting the

entire Kubernetes cluster (master + worker roles) within it.

4.2 Minikube Setup Steps

1. Install Prerequisites:

• Install Docker (e.g., Docker Desktop for Windows/Mac, or sudo apt

install docker.io for Linux).
• Install kubectl (Kubernetes CLI) to interact with the cluster.
• Install Minikube (e.g., via brew install minikube on macOS or curl
on Linux).

2. Start Minikube:
minikube start --driver=docker

This creates a single-node cluster using Docker as the driver.

3. Verify Cluster:
kubectl get nodes
minikube status

4. Access Dashboard (Optional):

minikube dashboard

5 Kubernetes Configuration
Kubernetes uses YAML files to define resources like Deployments, Services, Con-
figMaps, and StatefulSets. These files specify the desired state of the cluster,
which Kubernetes works to maintain.

3
5.1 Deployments for Stateless Apps
Deployments manage stateless applications (e.g., web apps). The replicas field
specifies the number of pod instances, not worker nodes. Example:
apiVersion: apps/v1
kind: Deployment
metadata:
name: web-app
spec:
replicas: 3
selector:
matchLabels:
app: web-app
template:
metadata:
labels:
app: web-app
spec:
containers:
- name: web-container
image: nanajanashia/k8s-demo-app:v1.0
ports:
- containerPort: 3000

The status section, automatically updated by Kubernetes, shows the current

state (e.g., availableReplicas: 3 if all pods are running).

5.2 StatefulSets for Stateful Apps

StatefulSets manage stateful applications (e.g., databases) with persistent storage
and unique identities. Example for MongoDB:
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: mongodb
spec:
serviceName: mongodb-service
replicas: 1
selector:
matchLabels:
app: mongodb
template:
metadata:
labels:
app: mongodb
spec:
containers:
- name: mongodb-container
image: mongo:latest
ports:
- containerPort: 27017

4
 volumeMounts:
- name: mongodb-data
mountPath: /data/db
volumeClaimTemplates:
- metadata:
name: mongodb-data
spec:
accessModes: [”ReadWriteOnce”]
resources:
requests:
storage: 1Gi

5.3 Services for Exposing Pods

Services provide a stable endpoint for accessing pods. They use selectors to route
traffic to pods with matching labels. Example:
apiVersion: v1
kind: Service
metadata:
name: web-service
spec:
selector:
app: web-app
ports:
- protocol: TCP
port: 80
targetPort: 3000
type: LoadBalancer

This exposes the web app externally on port 80, routing to port 3000 on the pods.

5.4 ConfigMaps and Secrets

• ConfigMap: Stores non-sensitive configuration data (e.g., URLs).

• Secret: Stores sensitive data (e.g., database credentials).

Example ConfigMap:
apiVersion: v1
kind: ConfigMap
metadata:
name: app-config
data:
mongodb-url: ”mongodb://mongodb-service:27017”

Example Secret:
apiVersion: v1
kind: Secret
metadata:
name: mongodb-secret

5
type: Opaque
data:
username: YWRtaW4= % base64 encoded
password: cGFzc3dvcmQ= % base64 encoded

6 Demo Project Structure

The demo project involves deploying a MongoDB database and a web application
in Kubernetes, with external access to the web app.

6.1 Project Files

• ConfigMap: Defines the MongoDB endpoint (e.g., mongodb://mongodb-service:27017

• Secret: Stores MongoDB username and password.

• StatefulSet and Service (MongoDB): Deploys MongoDB with persistent

storage and an internal ClusterIP service.

• Deployment and Service (Web App): Deploys the web app with a Load-
Balancer service for external access.

6.2 Key Considerations

• Use StatefulSet for MongoDB to ensure data persistence.

• Use Deployment for the web app, as it is stateless.

• Ensure the web app uses the ConfigMap and Secret to connect to MongoDB.

7 Troubleshooting Example
When pulling the nanajanashia/k8s-demo-app:v1.0 image, a MongoNetworkError
occurred due to a missing MongoDB service. Solution:
docker run --name mongodb -d mongo
docker run -p 3000:3000 --link mongodb:mongodb nanajanashia/k8s-demo
-app:v1.0

In Kubernetes, ensure the MongoDB service is running and accessible via the
ConfigMap.

8 Conclusion
Kubernetes is a powerful orchestration platform for managing containerized
applications. By understanding its components (master/worker nodes, kubelet,
pods) and configurations (Deployments, StatefulSets, Services, ConfigMaps, Se-
crets), you can deploy scalable and highly available applications. Minikube sim-
plifies local testing, and tools like kubectl provide precise control over the clus-
ter.