kube-no-trouble VS AKS

We also leverage tools like Kubent, popeye, kdave, and Pluto to help us manage API deprecations (when Kubernetes deprecates features in updates) and ensure the overall health of our infrastructure.

kubent has been my goto for this - you point it at your cluster, tell it the target version you want to use, and it'll let you know if you have any depreciated resources and what you'll need to change. It's simple to use, quick, and just does the job.

You have to constantly run tools like https://github.com/doitintl/kube-no-trouble / https://github.com/FairwindsOps/pluto.

A great tool for checking depreciations is kubent/kube-no-trouble: https://github.com/doitintl/kube-no-trouble

https://github.com/doitintl/kube-no-trouble can be used to check for deprecated/removed APIs - you'll need to fix these in your codebase. You should fix these before upgrading your cluster

Just upgraded a few clusters from 1.21 to 1.24 the past few weeks. Used kubent (Kube No Trouble https://github.com/doitintl/kube-no-trouble) before upgrading and reviewed the output. Pretty painless process all in all.

Ah yep, you're right. Just found someone asking the same thing from two days ago in the Azure AKS GitHub https://github.com/Azure/AKS/issues/3689. It should definitely be pulling the Ubuntu 22.04 fips compliant image if it were available.

Just did a search not sure related to your problem https://github.com/Azure/AKS/issues/85

As I could gather additional information which indicates some problems with the Monitoring Agent (high cpu on ama pods, no metrics delivered during CPU pressure, ama pods emitting logs), I opened a ticket on GitHub (https://github.com/Azure/AKS/issues/3469). Hopefully this helps to resolve the issue.

While I'm glad I saw this here, it's not one of our primary support channels. In the future, you can also raise an issue in our GitHub repository or file a support request.

You're dependent on the cloud provider for security settings in a lot of cases and they don't always get it right (e.g. AKS still doesn't have Node Authorization enabled! https://github.com/Azure/AKS/issues/3004)

For ASK even v1.22 got out of support yesterday 04 of December. https://github.com/Azure/AKS/releases/tag/2022-11-27

The Workload Identity add-on for AKS is still in progress. AKS has a public roadmap where you can find this kind of information.

Reference: https://github.com/Azure/AKS

For Thanos receive and query components to be available outside the cluster and secured with TLS, we will need ingress-nginx and cert-manager. For ingress, deploy the Helm chart using the following command, to account for this issue with AKS clusters >1.23: