Description of problem:
Some managed clusters attached to this ACM 2.7.2 se their governance-policy-framework-addon pod crashlooping, getting killed with an OOM
Version-Release number of selected component (if applicable):
2.7.2
How reproducible:
customer environment
Steps to Reproduce:
- deploy ACM 2.7.2
- attach clusters
- deploy policies
Actual results:
[...] containerStatuses: - containerID: cri-o://d54f011460f57c903914b82532d0cdb80aa3aabfac29110f137e3758710ce197 image: registry.redhat.io/rhacm2/acm-governance-policy-framework-addon-rhel8@sha256:b95517369a25f3ccda1d5d000e688ecf95b2403c9a38bd78ceac7cbf991254a6 imageID: registry.redhat.io/rhacm2/acm-governance-policy-framework-addon-rhel8@sha256:b95517369a25f3ccda1d5d000e688ecf95b2403c9a38bd78ceac7cbf991254a6 lastState: terminated: containerID: cri-o://d54f011460f57c903914b82532d0cdb80aa3aabfac29110f137e3758710ce197 exitCode: 137 finishedAt: "2023-03-20T10:30:59Z" reason: OOMKilled startedAt: "2023-03-20T10:30:26Z" name: governance-policy-framework-addon ready: false restartCount: 753 [...]
Expected results:
no crashloop or more information than "OOMKilled" available
Additional info:
the logs for the pod give no indication of any problem, it looks at first sight that the usage is just consuming enough memory to get into an OOM normally.
no procedure was found in the troubleshooting guide for this pod to remove or edit the limit, which could serve as a workaround.
- duplicates
-
ACM-3879 governance-policy-framework pod crashlooping on managed cluster
-
- Release Pending
-
