-
Bug
-
Resolution: Done
-
Normal
-
MCE 2.7.0
-
Quality / Stability / Reliability
-
2
-
False
-
None
-
False
-
-
-
-
SF Train-25, SF Train-26
-
Moderate
-
+
-
None
Description of problem:
After deleting a cluster, using ZTP GitOps, some resources, related to Observability, remains at the hub cluster. And this cannot be cleanly removed.:
$ oc get ns se..00152 -oyaml|grep message -A 2 message: All resources successfully discovered reason: ResourcesDiscovered status: "False" -- message: All legacy kube types successfully parsed reason: ParsedGroupVersions status: "False" -- message: All content successfully deleted, may be waiting on finalization reason: ContentDeleted status: "False" -- message: 'Some resources are remaining: manifestworks.work.open-cluster-management.io has 1 resource instances, rolebindings.authorization.openshift.io has 1 resource instances, rolebindings.rbac.authorization.k8s.io has 1 resource instances' -- message: 'Some content in the namespace has finalizers remaining: cluster.open-cluster-management.io/manifest-work-cleanup in 3 resource instances' reason: SomeFinalizersRemain $ oc get ns se..00152 NAME STATUS AGE se..00152 Terminating 5d21h
We can manually delete all the finalizers for the `manifestworks` and `rolebindings`, and it will finalize the cluster deletion.
In the past, something similar was solved by changing argocd deletion mode:
but not seems to be the same case. Actually, this fix was already included (saving many clusters deletion errors).
This seems to be a different case.
Version-Release number of selected component (if applicable):
ocp 4.14 and rhacm 2.10
How reproducible:
I cannot reproduce it. It happens few times, and we cannot find out the reason. We have a hub with hundreds of spokes, with many activity about creating/removing. So, this is happening like a few times by month.
Steps to Reproduce:
- ...
Actual results:
Expected results:
Additional info:
- clones
-
-
- Closed
-
- links to
-
RHSA-2025:145554
Multicluster Engine for Kubernetes 2.7.4 security updates
