Azure Kubernetes Service

Learning objectives#

• Create an AKS cluster attached to ACR
• Connect with kubectl
• Deploy a Deployment and a Service
• Scale pods and nodes

Explanation#

Kubernetes schedules containers (in pods) across a pool of VMs (nodes). AKS manages the control plane for you; you choose node sizes and counts.

A Deployment keeps a desired number of pod replicas running and handles rolling updates. A Service gives the pods a stable address; type LoadBalancer exposes it publicly.

Scale pods with kubectl scale or the Horizontal Pod Autoscaler; scale nodes with the cluster autoscaler.

AKS is powerful but demands Kubernetes skills. For simpler container hosting, consider Azure Container Apps first.

Code example: Deploying an API to AKS#

1# az aks create -g rg-shop -n aks-shop --node-count 2 --generate-ssh-keys --attach-acr acrshop12345
2# az aks get-credentials -g rg-shop -n aks-shop
3apiVersion: apps/v1
4kind: Deployment
5metadata: { name: shop-api }
6spec:
7  replicas: 3
8  selector: { matchLabels: { app: shop-api } }
9  template:
10    metadata: { labels: { app: shop-api } }
11    spec:
12      containers:
13        - name: api
14          image: acrshop12345.azurecr.io/shop-api:1.0.0
15          ports: [{ containerPort: 8080 }]
16          resources: { requests: { cpu: 100m, memory: 128Mi }, limits: { memory: 256Mi } }
17---
18apiVersion: v1
19kind: Service
20metadata: { name: shop-api }
21spec:
22  type: LoadBalancer
23  selector: { app: shop-api }
24  ports: [{ port: 80, targetPort: 8080 }]
25# kubectl apply -f shop-api.yaml && kubectl get svc shop-api

NAME       TYPE           CLUSTER-IP    EXTERNAL-IP     PORT(S)
shop-api   LoadBalancer   10.0.142.17   20.xxx.xxx.xx   80:31540/TCP

This example is illustrative. Copy or download it and run it in your own environment; the expected output is shown above.

Step-by-step explanation#

1. --attach-acr lets the cluster pull from your registry.
2. get-credentials configures kubectl for the cluster.
3. The Deployment runs three replicas with resource requests and limits.
4. The LoadBalancer Service gets a public IP and forwards port 80 to the containers’ 8080.

Real-world use case#

Common mistakes#

Best practices#

• Set resource requests and limits
• Use managed identities and Entra integration
• Enable the cluster autoscaler
• Use an ingress controller with TLS for HTTP apps

Remember this#

Practice exercise#

1kubectl scale deployment shop-api --replicas=5
2kubectl get pods -l app=shop-api -w

Mini quiz#

Want more? Practise all Microsoft Azure MCQs · Microsoft Azure interview questions

Related topics#

My notes#

Saved in this browser.