I'm looking for some tips on how to transition Helm releases into ArgoCD while ensuring we don't end up with orphaned Kubernetes resources. What are some techniques or hooks you've used to make this migration smooth and efficient?
1 Answer
That's a valid concern, especially in more complex setups. If you're migrating a single Application object and you can ensure all names match, it gets easier. But for environments where everything is based on AppSets—which generate Applications with various labels and names—it can get tricky. You might need to adjust the nameoverride or fullnameoverride in your values files, but that can create messes, especially with nested charts.
We're migrating around a hundred services from Helm CLI over to ArgoCD, and I’d suggest taking small outages and removing the old install before deploying the new one. However, this approach might leave behind some resources, particularly those created by operators that don't self-clean. We plan to handle that semi-manually by tagging ArgoCD-managed resources distinctly so we can identify and clean them up later. Lots of work ahead!

Related Questions
Can't Load PhpMyadmin On After Server Update
Redirect www to non-www in Apache Conf
How To Check If Your SSL Cert Is SHA 1
Windows TrackPad Gestures