-
Bug
-
Resolution: Unresolved
-
Minor
-
None
-
ACM 2.11.Z
-
2
-
False
-
None
-
False
-
-
-
Observability Sprint 32, Observability Sprint 33, Observability Sprint 34, Observability Sprint 35, Obs Sprint 36, Observability Sprint 37, Observability Sprint 38
-
None
Expected Behavior:
I am able to set an CPU limit of my choosing or none at all.
Actual behavior:
a CPU limit value of `'0'` is injected into the `observability-thanos-receive-default` STS
Reproduction Steps:
- Modify the MCO spec with the following patch:
apiVersion: observability.open-cluster-management.io/v1beta2 kind: MultiClusterObservability metadata: name: observability spec: advanced: receive: resources: requests: cpu: 300m memory: 768Mi limits: memory: 768Mi
or
apiVersion: observability.open-cluster-management.io/v1beta2 kind: MultiClusterObservability metadata: name: observability spec: advanced: receive: resources: requests: cpu: 300m memory: 768Mi limits: cpu: 300m memory: 768Mi
- wait for MCO to be reconciled
- Observe that on the observability-thanos-receive-default StatefulSet that `spec.template.containers[0].resources.limits.cpu` is set to the string '0' instead of null in the case of the first patch or '300m' in the second.
This seems to be a bug that was introduced in 2.11 as I can reproduce this in 2.11 but not 2.10.