This repository has been archived by the owner on May 6, 2022. It is now read-only.
Service Catalog controller-manager fails to start when unrelated API server is unavailable #2418
Labels
kind/bug
Categorizes issue or PR as related to a bug.
Comments
Merged
10 tasks
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Bug Report
What happened:
Service Catalog controller-manager is failing to start when some unrelated API server is down.
This is caused by a change kubernetes/kubernetes#66932 that makes discovery return details about unavailable resources.
https://github.com/kubernetes-incubator/service-catalog/blob/master/cmd/controller-manager/app/controller_manager.go#L280-L283
What you expected to happen:
Service Catalog controller-manager should explicitly check availability of "servicecatalog.k8s.io" resources is failing to start when some unrelated API server is down.
We should fix the code to ignore errors in discovery unrelated to "servicecatalog.k8s.io" resource group.
How to reproduce it (as minimally and precisely as possible):
Register a new
APIServicewithout starting an appropriate API server.Deploy Service Catalog and see controller-manager being continuously restarted.
The text was updated successfully, but these errors were encountered: