Track 7: Azure Kubernetes Service (AKS)

This is the capstone track of the whole curriculum. You've done Linux, Docker (including pushing to ACR and deploying to ACI/App Service), and Kubernetes on a local cluster (kind/minikube) — then took a deliberate detour through general networking, Azure networking, and Azure Container Apps. This track takes everything you know about kubectl, Deployments, Services, ConfigMaps/Secrets, Ingress, HPA, and Helm, and points it at a real, managed, billable Kubernetes cluster in Azure — the hand-operated alternative to the Container Apps platform that was quietly managing all of this for you last track. Where Container Apps abstracted away the cluster, the node pools, and the Environment's networking, AKS hands all of that back to you — which is exactly the trade-off the previous track's capstone left you asking about.

Cost warning: every module that creates an AKS cluster should be cleaned up with az group delete when you're done for the day. An idle AKS cluster still bills for its worker node VMs (and their disks, and any load balancers you created) even if you aren't running any workloads on it. Nothing here is prohibitively expensive if you clean up promptly, but leaving a cluster running over a weekend by accident is a real way to get a real bill. When in doubt: az group delete --name <rg> --yes --no-wait.

How this track works

  • Go in order — module 01 assumes module 00 is done, module 03 assumes you can already deploy a Pod from module 02, and so on.
  • Every module (except this index and the capstone) follows the same shape: Why this matters → Concepts → Command reference → Hands-on exercises → Common mistakes & troubleshooting → Checkpoint quiz → Next.
  • All exercises use real az and kubectl commands against your actual Azure subscription. There is no local-only AKS simulator — this is the point of the track.
  • Each module's exercises end with an explicit cleanup step, or explicitly tell you the cleanup is deferred to a later exercise in the same module. Don't skip these.
  • Module 09 is a capstone project with no quiz — it asks you to combine everything from modules 00-08 (and the Helm chart from the Kubernetes track capstone) into one real, working, monitored, autoscaling, CI/CD-deployed application on AKS.

Modules

# Module What it covers Rough time
00 azure-cli-and-prerequisites Installing/updating the Azure CLI, az login, subscriptions, kubectl install, resource groups, AKS pricing model 45-60 min
01 creating-an-aks-cluster az aks create, node pools, cluster tiers, get-credentials, verifying nodes, cleanup 60-90 min
02 deploying-workloads-to-aks Deployments, Services, ConfigMaps/Secrets on a real cluster; what's different from kind/minikube 60-90 min
03 acr-integration-with-aks az aks update --attach-acr, managed identity image pulls, image pull secrets as a fallback 45-60 min
04 aks-networking-loadbalancer-and-ingress LoadBalancer Services, public IPs, ingress-nginx / App Routing add-on, DNS 60-90 min
05 scaling-aks-cluster-autoscaler-and-hpa Cluster Autoscaler, node pools scaling, combining with HPA from the Kubernetes track 60-90 min
06 monitoring-aks-azure-monitor-container-insights Container Insights, Log Analytics, Azure Monitor metrics and alerts 60 min
07 security-aks-aad-rbac-and-keyvault Azure AD-integrated Kubernetes RBAC, managed identities, Key Vault secrets via CSI driver 75-90 min
08 cicd-github-actions-to-aks GitHub Actions building images, pushing to ACR, rolling out to AKS 60-90 min
09 capstone-project End-to-end project: cluster, ACR, Helm deploy, Ingress, monitoring, HPA under load, CI/CD 3-6 hours

Prerequisites

  • An active Azure subscription (already confirmed for this curriculum).
  • Everything from 03-kubernetes: comfortable with kubectl, Deployments, Services, ConfigMaps/Secrets, Ingress, HPA, and Helm on a local cluster.
  • Everything from 02-docker: building images, pushing to Azure Container Registry, and basic ACI/App Service deploys.

Back to main curriculum