Build, test, and deploy containers to Azure Kubernetes Service (AKS) using GitHub Actions

GitHub Actions gives you the flexibility to build an automated software development lifecycle workflow. You can use multiple Kubernetes actions to deploy to containers from Azure Container Registry (ACR) to Azure Kubernetes Service (AKS) with GitHub Actions.

Prerequisites

• An Azure account with an active subscription. Create a trial subscription.
• A GitHub account. If you don't have one, sign up for free.
  • When using GitHub Actions, you need to configure the integration between Azure and your GitHub repository.
• An existing AKS cluster with an attached ACR. If you don't have one, see Authenticate with ACR from AKS.

GitHub Actions for AKS

With GiHub Actions, you can automate your software development workflows from within GitHub. For more information, see [GitHub Actions for Azure][github-actions].

The following table lists the available actions for AKS:

Use GitHub Actions with AKS

As an example, you can use GitHub Actions to deploy an application to your AKS cluster every time a change is pushed to your GitHub repository. This example uses the Azure Vote application.

Fork and update the repository

1. Navigate to the Azure Vote repository and select Fork.

2. Update the azure-vote-all-in-one-redis.yaml to use your ACR for the azure-vote-front image. Replace <registryName> with the name of your registry.

   ...
         containers:
         - name: azure-vote-front
           image: <registryName>.azurecr.cn/azuredocs/azure-vote-front:v1
   ...
   

3. Commit the updated azure-vote-all-in-one-redis.yaml to your repository.

Create secrets

1. Create a service principal to access your resource group with the Contributor role using the [az ad sp create-for-rbac][az-ad-sp-create-for-rbac] command. Replace <SUBSCRIPTION_ID> with the subscription ID of your Azure account and <RESOURCE_GROUP> with the name of the resource group containing your ACR.

   az ad sp create-for-rbac \
       --name "ghActionAzureVote" \
       --scope /subscriptions/<SUBSCRIPTION_ID>/resourceGroups/<RESOURCE_GROUP> \
       --role Contributor \
       --json-auth
   

   Your output should look similar to the following example output:

   {
     "clientId": <clientId>,
     "clientSecret": <clientSecret>,
     "subscriptionId": <subscriptionId>,
     "tenantId": <tenantId>,
     ...
   }
   

2. Navigate to your GitHub repository settings and select Security > Secrets and variables > Actions.

3. For each secret, select New Repository Secret and enter the name and value of the secret.

   Secret name	Secret value
   AZURE_CREDENTIALS	The entire JSON output from the az ad sp create-for-rbac command.
   service_principal	The value of <clientId>.
   service_principal_password	The value of <clientSecret>.
   subscription	The value of <subscriptionId>.
   tenant	The value of <tenantId>.
   registry	The name of your registry.
   repository	azuredocs
   resource_group	The name of your resource group.
   cluster_name	The name of your cluster.

For more information about creating secrets, see Encrypted Secrets.

Create actions file

1. In your repository, create a .github/workflows/main.yml and paste in the following contents:

   name: build_deploy_aks
   on:
     push:
       paths:
         - "azure-vote/**"
   jobs:
     build:
       runs-on: ubuntu-latest
       steps:
         - name: Checkout source code 
           uses: actions/checkout@v3
         - name: ACR build
           id: build-push-acr
           uses: azure/acr-build@v1
           with:
             service_principal: ${{ secrets.service_principal }}
             service_principal_password: ${{ secrets.service_principal_password }}
             tenant: ${{ secrets.tenant }}
             registry: ${{ secrets.registry }}
             repository: ${{ secrets.repository }}
             image:  azure-vote-front
             folder: azure-vote
             branch: master
             tag: ${{ github.sha }}
         - name: Azure login
           id: login
           uses: azure/login@v1.4.3
           with:
             creds: ${{ secrets.AZURE_CREDENTIALS }}
         - name: Set AKS context
           id: set-context
           uses: azure/aks-set-context@v3
           with:
             resource-group: '${{ secrets.resource_group }}' 
             cluster-name: '${{ secrets.cluster_name }}'
         - name: Setup kubectl
           id: install-kubectl
           uses: azure/setup-kubectl@v3
         - name: Deploy to AKS
           id: deploy-aks
           uses: Azure/k8s-deploy@v4
           with:
             namespace: 'default'
             manifests: |
                azure-vote-all-in-one-redis.yaml
             images: '${{ secrets.registry }}.azurecr.cn/${{ secrets.repository }}/azure-vote-front:${{ github.sha }}'
             pull-images: false 
   

   The on section contains the event that triggers the action. In the example file, the action triggers when a change is pushed to the azure-vote directory.

   The steps section contains each distinct action:

   1. Checkout source code uses the GitHub Actions Checkout Action to clone the repository.
   2. ACR build uses the Azure Container Registry Build Action to build the image and upload it to your registry.
   3. Azure login uses the Azure Login Action to sign in to your Azure account.
   4. Set AKS context uses the Azure AKS Set Context Action to set the context for your AKS cluster.
   5. Setup kubectl uses the Azure AKS Setup Kubectl Action to install kubectl on your runner.
   6. Deploy to AKS uses the Azure Kubernetes Deploy Action to deploy the application to your Kubernetes cluster.

2. Commit the .github/workflows/main.yml file to your repository.

3. To confirm the action is working, update the azure-vote/azure-vote/config_file.cfg with the following contents:

   # UI Configurations
   TITLE = 'Azure Voting App'
   VOTE1VALUE = 'Fish'
   VOTE2VALUE = 'Dogs'
   SHOWHOST = 'false'
   

4. Commit the updated azure-vote/azure-vote/config_file.cfg to your repository.

5. In your repository, select Actions and confirm a workflow is running. Then, confirm the workflow has a green checkmark and the updated application is deployed to your cluster.

Next steps

Review the following starter workflows for AKS. For more information, see Using starter workflows.

• Azure Kubernetes Service (Basic)
• Azure Kubernetes Service Helm
• Azure Kubernetes Service Kustomize
• Azure Kubernetes Service Kompose