-
Story
-
Resolution: Unresolved
-
Critical
-
None
The test 'test_must_gather_and_vm_same_node' is permafailing when it's run on a non-compact cluster (3 masters + 3 workers).
The reason is that the must-gather cannot be scheduled as it has nodeAfinnity to run on masters:
spec:
affinity:
nodeAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: kubernetes.io/hostname
operator: In
values:
- c01-rlobillo-420-qxhkq-master-0
- c01-rlobillo-420-qxhkq-master-2
- c01-rlobillo-420-qxhkq-master-1
and nodeName attribute set to the worker where the VM is running. Therefore, it is never scheduled and killed with 137 error:
2025-10-08T19:13:11.051651 utilities.must_gather ERROR must-gather raised the following error: openshift-must-gather-pmr5w/must-gather-v2xqw unexpectedly terminated: exit code: 137, reason: ContainerStatusUnknown, message: The container could not be located when the pod was terminated
The test was passing until https://github.com/RedHatQE/openshift-virtualization-tests/pull/705 fixed it and started to fail in non-compact clusters.
Solution to be decided.
- clones
-
-
- In Progress
-