-
Bug
-
Resolution: Done-Errata
-
Undefined
-
None
-
4.12.z
-
Moderate
-
No
-
CNF RAN Sprint 240
-
1
-
False
-
-
Description of problem:
The docs say "To trigger auto-creation of the ClusterGroupUpgrade CR, remove the ztp-done label from the cluster and delete the empty ClusterGroupUpgrade CR that was previously created in the ztp-install namespace." It seems that the order is important. If you delete the "ztp-done" label, and then delete the CGU, the operator will reconcile and create a new CGU. However, if you delete the CGU, and then the "ztp-done" label, nothing happens, a new CGU is not created. This is because the operator only watches for CGU deletions to trigger a reconcile. It does not watch the ManagedCluster for updates. So if you delete the CGU, a reconcile happens (with no action, since the cluster has the "ztp-done" label), but when "ztp-done" is removed from the ManagedCluster, no reconcile happens. Manually deleting the controllermanager pod for ManagedClusterForCgu causes the reconcile to happen, and a new CGU to be created.
Version-Release number of selected component (if applicable):
TALM 4.12.4
How reproducible:
Steps to Reproduce:
1. 2. 3.
Actual results:
Expected results:
Additional info:
- blocks
-
OCPBUGS-17755 Deleting "ztp-done" label after deleting CGU doesn't trigger reconcile
- Closed
-
OCPBUGS-17756 Deleting "ztp-done" label after deleting CGU doesn't trigger reconcile
- Closed
- is cloned by
-
OCPBUGS-17755 Deleting "ztp-done" label after deleting CGU doesn't trigger reconcile
- Closed
-
OCPBUGS-17756 Deleting "ztp-done" label after deleting CGU doesn't trigger reconcile
- Closed
- links to
-
RHEA-2023:112754 OpenShift Container Platform 4.14.0 CNF vRAN extras update
-
RHEA-2023:116214 OpenShift Container Platform 4.13.9 CNF vRAN extras update
- mentioned on