Description of problem:
In IBM Cloud managed environment due to missing serving cert volume mount, the operator is trying to create the cert in the tmp folder, but the readRootFileSystem is enabled so it can't do that. This causes the operator to fail with a /tmp readonly error.
Version-Release number of selected component (if applicable):
4.21+
How reproducible:
Always
Steps to Reproduce:
1.Use hypershift 4.21-rc-1 version and deploy a HostedCluster
2.Check cluster-csi-snapshot-controller-operator pod logs
Actual results:
cluster-csi-snapshot-controller-operator pods in CrashLoopBackoff state
Expected results:
cluster-csi-snapshot-controller-operator works properly, creating the serving cert
Additional info:
│ W0114 12:49:26.197738 1 cmd.go:257] Using insecure, self-signed certificates │ │ F0114 12:49:26.198152 1 cmd.go:182] mkdir /tmp/serving-cert-824191421: read-only file system │ │ stream closed: EOF for openshift-cluster-storage-operator/csi-snapshot-controller-operator-8678c47545-6k59s (csi-snapshot-controller-operator)
- blocks
-
OCPBUGS-74301 csi-snapshot-controller-operator fails to start when placed on HCP guest cluster
-
- POST
-
- is cloned by
-
OCPBUGS-74301 csi-snapshot-controller-operator fails to start when placed on HCP guest cluster
-
- POST
-
- links to