Deploy your first Service to Azure Container Services (AKS)

In this post I’ll show you how to Deploy your first Service to Azure Container Services (AKS).

Prerequisites:

• Azure CLI installed and basic knowledge experience.
• Docker installed and basic knowledge.
• Azure Subscription
• Kubernetes experience.

Create a resource group:

Firt create a Resource Group. Be aware that at the time of writing AKS is not available in all Azure regions.

az group create -l westeurope -n aks

Create an Azure Container Registry:

Probably you’ll want a private registry to upload your docker images, so let’s create an Azure Container Registry instance.

az acr create -g aks -n myregistry --sku Basic --admin-enabled

Login to the Azure Container Registry:

You can login to the Container Registry using the following Azure CLI command:

az acr login -n myregistry

Or you can get the Key and login with docker:

az acr credential show -n myregistry
docker login myregistry.azurecr.io -u myregistry -p [PASSWORD FROM PREVIOUS COMMAND]

Push an image to the Azure Container Registry:

In this step we are going to pull an image from docker hub, and then upload it to the Container Registry created in step 2. Feel free to use your own docker image with a working web application.

docker pull yaros1av/hello-core
docker tag yaros1av/hello-core myregistry.azurecr.io/hello-core:1.0
docker push myregistry.azurecr.io/hello-core:1.0

Create AKS cluster

Now create the AKS cluster:

az aks create -g aks --name myAKSCluster --generate-ssh-keys

Install kubectl

To manage the cluster you’ll need to install kubectl so run the following command:

az aks install-cli

Get cluster credentials

Get the cluster credentials and check the connectivity so you can start working with the AKS cluster.

az aks get-credentials -g aks -n myAKSCluster
kubectl get nodes

Create a Secret to hold the registry credentials.

You have to add the Azure Container Registry credentials to your AKS service in order to be able to pull the images. This is acomplished by creating a secret:

kubectl create secret docker-registry [SECRET NAME] --docker-server=myregistry.azurecr.io --docker-username=myregistry --docker-password=[THE REGISTRY PASWORD FROM STEP 3] --docker-email=[EMAIL ADDRESS]

Check for the secret:

kubectl describe secret

Create a kubernetes deployment definition.

Create a file [your deployment].yml with the following contents (replace the secret name):

apiVersion: apps/v1beta1
kind: Deployment
metadata:
  name: my-api
spec:
  replicas: 1
  template:
    metadata:
      labels:
        app: my-api
    spec:
      containers:
      - name: my-api
        image: myregistry.azurecr.io/hello-core:1.0
        ports:
        - containerPort: 80
      imagePullSecrets:
      - name: [SECRET NAME]
---
apiVersion: v1
kind: Service
metadata:
  name: my-api
spec:
  type: LoadBalancer
  ports:
  - port: 80
  selector:
    app: my-api

As you can see the file references the image pushed to the Container Service with the secret created in previous steps.

Deploy your service.

Run the following comnmand to deploy your service to AKS:

kubectl create -f [your deployment].yml

Get the puplic IP for your service

To try your service you’ll need to get the public IP (It can take a while).

kubectl.exe get service/my-api -w

Once you get the IP go ahead and test the service.

Update your deployment to different version of the image

If you need to update your service to use another version of the dokcer image run the following command:

kubectl set image deployment/my-api [image name]=myregistry.azurecr.io/[image name]:[new version]

Add an autoscale rule

To add an autoscale rule to your service, run the following command:

kubectl autoscale deployment my-api --min=2 --max=5 --cpu-percent=80

Hope it helps!

• ←Previous
• Next→