Akash Roy

Posted on Oct 8

Deploy to Azure Kubernetes Service (AKS) using Azure Pipelines

#azure #devops #kubernetes #cicd

If you’ve already built your first Azure Pipeline, this next step will show you how to take your code all the way to production — deploying automatically to an Azure Kubernetes Service (AKS) cluster.

This post is a simple walkthrough for anyone who wants to see their containerized app running on Azure, without deep Kubernetes expertise.

What We’ll Do

Containerize a sample app
Push the image to Azure Container Registry (ACR)
Create a YAML-based pipeline to deploy it on AKS
Validate the deployment

Step 1: Prepare Your App and ACR

Let’s assume you have a simple Node.js or Python app.

Create a Dockerfile in your project root:

FROM node:18
WORKDIR /app
COPY . .
RUN npm install
CMD ["npm", "start"]

Now, build and push this image to Azure Container Registry (ACR):

az acr build --registry <your-acr-name> --image sampleapp:v1 .

Step 2: Connect Azure DevOps with ACR and AKS

In your Azure DevOps project:

Go to Project Settings → Service Connections
Create two connections:
- Azure Resource Manager (for AKS)
- Docker Registry (for ACR)

These connections will authenticate your pipeline securely.

Step 3: Create a Deployment YAML

Add deployment.yaml in your repo:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: sampleapp
spec:
  replicas: 2
  selector:
    matchLabels:
      app: sampleapp
  template:
    metadata:
      labels:
        app: sampleapp
    spec:
      containers:
      - name: sampleapp
        image: <your-acr-name>.azurecr.io/sampleapp:v1
        ports:
        - containerPort: 80

Step 4: Build the Azure Pipeline

Create .azure-pipelines/aks-deploy.yaml:

trigger:
- main

pool:
  vmImage: ubuntu-latest

variables:
  imageName: sampleapp
  tag: '$(Build.BuildId)'

stages:
- stage: Build
  jobs:
  - job: BuildImage
    steps:
    - task: Docker@2
      inputs:
        containerRegistry: 'acr-connection'
        repository: '$(imageName)'
        command: 'buildAndPush'
        Dockerfile: '**/Dockerfile'
        tags: |
          $(tag)

- stage: Deploy
  dependsOn: Build
  jobs:
  - job: DeployToAKS
    steps:
    - task: KubernetesManifest@1
      inputs:
        action: deploy
        kubernetesServiceConnection: 'aks-connection'
        manifests: '**/deployment.yaml'
        containers: |
          $(imageName)=$(imageName):$(tag)

Commit and push this YAML file to your main branch.

Step 5: Run and Verify

Once the pipeline completes, verify the deployment:

kubectl get pods
kubectl get svc

You’ll see your app live on AKS — deployed automatically through Azure DevOps!

Wrapping Up

You’ve just connected CI/CD to a live Kubernetes cluster. From here, you can:

Add health checks or rolling updates
Integrate Helm for versioned deployments
Set up Blue-Green or Canary strategies

This forms the core of a modern cloud deployment pipeline — repeatable, automated, and secure.

Next up: I’ll write about securing pipelines using Azure Key Vault for secret management — stay tuned.

DEV Community