-
Bug
-
Resolution: Done-Errata
-
Major
-
None
-
4.18.z
-
Quality / Stability / Reliability
-
False
-
-
None
-
Important
-
No
-
None
-
None
-
CNF Compute Sprint 268, CNF Compute Sprint 269
-
2
-
In Progress
-
Release Note Not Required
-
None
-
None
-
None
-
None
-
None
This is a clone of issue OCPBUGS-51261. The following is the description of the original issue:
—
Description of problem:
updates on NROP CR that lead to point to different nodes and node groups are not reflected under CM even when kubeletconfig that targets the new node group is there.
Version-Release number of selected component (if applicable):
all
How reproducible:
always
Steps to Reproduce:
cluster setup: 2 mcps targeting worker nodes. 2 kubeletconfigs one per mcp from above, and each with different config (TM scope and policy) 1. create NROP CR with node group pointing to mcp-1 2. update the CR node group to point to mcp-2
Actual results:
the new NRTs are created but with the first config for mcp-1 and that is because no new CM is created.
Expected results:
New CM should be created and the NRTs should retrieve the config data from the new config
Additional info:
workaround is to poke the kubeletconfig objects. any update there addition or deleton will trigger the right behavior
- blocks
-
OCPBUGS-53155 [4.17] NROP controller doesn't create new CM on node groups updates
-
- Closed
-
- is blocked by
-
OCPBUGS-51261 NROP controller doesn't create new CM on node groups updates
-
- Closed
-
- links to
-
RHEA-2025:147938 OpenShift Container Platform 4.18.Z low-latency extras update
- mentioned on
(2 mentioned on)