Using labels for service routing in Kubernetes - Mini Project: Build & Apply

Using labels for service routing

📖 Scenario: You are managing a Kubernetes cluster that runs multiple versions of a web application. You want to route traffic to the correct version using labels and services.

🎯 Goal: Create two deployments with different labels and a service that routes traffic to one of them using label selectors.

📋 What You'll Learn

Create a deployment named web-v1 with label version: v1

Create a deployment named web-v2 with label version: v2

Create a service named web-service that routes traffic to pods with label version: v1

Use correct Kubernetes YAML syntax for deployments and service

Verify the service routes to the correct pods by checking labels

💡 Why This Matters

🌍 Real World

In real Kubernetes clusters, labels and selectors help route traffic to the right pods, enabling blue-green deployments and canary releases.

💼 Career

Understanding labels and service routing is essential for Kubernetes administrators and DevOps engineers to manage application versions and traffic flow.

1

Create the first deployment with label

Create a deployment named web-v1 with label version: v1 and container image nginx:1.19. Use 2 replicas.

Kubernetes

# Create deployment web-v1 with label version: v1
# Your code here

Need a hint?

Use metadata.name for deployment name, spec.replicas for number of pods, and metadata.labels inside template for pod labels.

2

Create the second deployment with a different label

Create a deployment named web-v2 with label version: v2 and container image nginx:1.20. Use 2 replicas.

Kubernetes

apiVersion: apps/v1
kind: Deployment
metadata:
  name: web-v1
spec:
  replicas: 2
  selector:
    matchLabels:
      version: v1
  template:
    metadata:
      labels:
        version: v1
    spec:
      containers:
      - name: nginx
        image: nginx:1.19

# Create deployment web-v2 with label version: v2
# Your code here

Need a hint?

Follow the same structure as web-v1 deployment but change the name, label, and image.

3

Create a service to route traffic using label selector

Create a service named web-service that routes traffic to pods with label version: v1. Use port 80 for the service and target port 80 for the pods.

Kubernetes

apiVersion: apps/v1
kind: Deployment
metadata:
  name: web-v1
spec:
  replicas: 2
  selector:
    matchLabels:
      version: v1
  template:
    metadata:
      labels:
        version: v1
    spec:
      containers:
      - name: nginx
        image: nginx:1.19
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: web-v2
spec:
  replicas: 2
  selector:
    matchLabels:
      version: v2
  template:
    metadata:
      labels:
        version: v2
    spec:
      containers:
      - name: nginx
        image: nginx:1.20

# Create service web-service with selector version: v1
# Your code here

Need a hint?

Use kind: Service with spec.selector matching version: v1 and define ports correctly.

4

Verify the service routes to the correct pods

Run the command kubectl get pods --selector=version=v1 --output=wide and print the output to verify pods with label version: v1 are selected.

Kubernetes

apiVersion: apps/v1
kind: Deployment
metadata:
  name: web-v1
spec:
  replicas: 2
  selector:
    matchLabels:
      version: v1
  template:
    metadata:
      labels:
        version: v1
    spec:
      containers:
      - name: nginx
        image: nginx:1.19
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: web-v2
spec:
  replicas: 2
  selector:
    matchLabels:
      version: v2
  template:
    metadata:
      labels:
        version: v2
    spec:
      containers:
      - name: nginx
        image: nginx:1.20
---
apiVersion: v1
kind: Service
metadata:
  name: web-service
spec:
  selector:
    version: v1
  ports:
  - protocol: TCP
    port: 80
    targetPort: 80

# Run kubectl command to get pods with label version=v1
# Your code here

Need a hint?

Use print() to simulate the output of kubectl get pods --selector=version=v1 --output=wide.