-
Bug
-
Resolution: Unresolved
-
Major
-
None
-
4.22.0, 4.21.z
-
None
Description of problem:
[groupsnapshot_controller] race condition in group snapshot creation with incomplete status
Version-Release number of selected component (if applicable):
4.22.0-0.nightly-2026-01-21-055002
How reproducible:
Randomly (about 1%)
Steps to Reproduce:
1. Install a openshift cluster with vgs feature gate enabled.
2. Create pvcs with volumeBindingMode: Immediate storageclass and label them with app.kubernetes.io/name: myapp-data.
3. Create a volumegroupsnapshot for the group volumes.
4. Wait for the volumegroupsnapshot ready to use check its content status the volumeSnapshotInfoList should have full info.
Actual results:
In step 4 the the common controller processes VolumeGroupSnapshotContent with incomplete VolumeSnapshotInfoList data which left empty volumeHandle and snapshotHandle fields even if the VolumeGroupSnapshot status is READYTOUSE: true.
Expected results:
In step 4 the content status the volumeSnapshotInfoList should have full info
Additional info:
CI flakes: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/periodic-ci-openshift-release-master-ci-4.22-e2e-aws-ovn-techpreview/2013852668694368256 Upstream issue: https://github.com/kubernetes-csi/external-snapshotter/issues/1357 Mock reproducer: https://github.com/jsafrane/external-snapshotter/commit/5a2328df0ec5263c230092a209f9de1134ff1a6a