This challenge will cover the basics of Azure Arc for Kubernetes, as well as Azure Arc for Data Services.
You need to fulfill these requirements to complete this challenge:
- Deploy a non-AKS Kubernetes cluster somewhere (it could be on Azure or anywhere else) and connect it with Azure so that it appears as an ARM resource in the portal. It is recommended using VM sizes for the worker nodes with at least 4 cores and premium disk support, since Arc for data is quite demanding in terms of resources.
- Configure an Azure Policy to prevent privileged containers from running in the cluster.
- Use Gitops to deploy an application to the cluster. You can take the application we have been using in this Hack (the web/api container images) or any other container image
- Deploy an Arc enabled database in the new cluster. Make sure the policy previously deployed does not interfere with the Arc data controller, in case it needs to run privileged containers.
- Show an Arc enabled non-AKS cluster
- An application has been deployed with Gitops
- Participants can demonstrate how Azure Policies prevent the usage of privileged containers
- An Arc enabled database is deployed in the cluster
- Connect your AKS cluster from previous challenges to the same Gitops repo as the Arc enabled cluster, and make sure that changes to the application are propagated to both clusters
- Explain how you would distribute application secrets via GitOps
These docs might help you achieving these objectives: