-
Bug
-
Resolution: Unresolved
-
Critical
-
4.20.0, 4.21.0
-
None
-
Quality / Stability / Reliability
-
False
-
-
None
-
Critical
-
None
-
None
-
Proposed
-
None
-
In Progress
-
Release Note Not Required
-
None
-
None
-
None
-
None
-
None
This is a clone of issue OCPBUGS-61499. The following is the description of the original issue:
—
Description of problem:
A limit was added for kube-controller-manager resources. Without taking limits into consideration when dynamically setting resource requests we can prevent the kube-controller-manager from getting deployed.
Version-Release number of selected component (if applicable):
4.20 and later
How reproducible:
Always
Steps to Reproduce:
1. Create a HostedCluster with the following annotation to size the KubeControllerManager:
resource-request-override.hypershift.openshift.io/kube-controller-manager.kube-controller-manager: memory=2560Mi
2. Wait for the HostedCluster to provision
Actual results:
Nodes never join the cluster because the kube controller manager fails to run.
Expected results:
Nodes join the cluster
Additional info:
The kube controller manager fails to reconcile with the following error: - lastTransitionTime: "2025-09-09T22:05:40Z" message: 'Deployment.apps "kube-controller-manager" is invalid: spec.template.spec.containers[0].resources.requests: Invalid value: "2560Mi": must be less than or equal to memory limit of 1Gi' reason: ReconciliationError status: "False" type: RolloutComplete
- clones
-
-
- Verified
-
- is blocked by
-
-
- Verified
-
- links to
