Created attachment 1907161 [details] Testlog Description of problem (please be detailed as possible and provide log snippests): Restoring snapshot fails with size constraint when the parent pvc is expanded. The restore operation succeeds when the restore pvc is of the same size as the expanded parent pvc. This test case is executed as part of ocs-ci tier2 test tests/manage/pv_services/test_expansion_snapshot_clone.py::TestExpansionSnapshotClone::test_expansion_snapshot_clone[thin-thin] Version of all relevant components (if applicable): ODF : 4.11.0-137 Does this issue impact your ability to continue to work with the product (please explain in detail what is the user impact)? Is there any workaround available to the best of your knowledge? Rate from 1 - 5 the complexity of the scenario you performed that caused this bug (1 - very simple, 5 - very complex)? Can this issue reproducible? Yes Can this issue reproduce from the UI? If this is a regression, please provide more details to justify this: Steps to Reproduce: 1. Create a pvc of 2Gi, expand it to 4Gi 2. Create a volume snapshot of the 4Gi pvc 3. Verify that the snapshot "READY TO USE" state is "true" 4. Expand parent pvc to 6GiB 5. Restore snapshot to a pvc of 4Gi Actual results: Restore of snapshot fails with the following error Warning ProvisioningFailed 2s (x7 over 34s) openshift-storage.cephfs.csi.ceph.com_csi-cephfsplugin-provisioner-7 67f4b7584-zvxpr_62f81ad9-e73f-4fd3-b364-65ae3e858533 failed to provision volume with StorageClass "ocs-storagecluster- cephfs": rpc error: code = InvalidArgument desc = cannot restore from snapshot 0001-0011-openshift-storage-000000000000 0001-2e74dff5-22f0-11ed-ab35-0a580a800214: volume size 6442450944 is smaller than source volume size 4294967296 Expected results: Restore of snapshot should be successful Additional info: Must-gather- https://drive.google.com/file/d/1YPQnacblJKwtY1jtE43b2Kf1qGxK7a0R/view?usp=sharing
This is a regression caused by an upstream fix https://github.com/ceph/ceph-csi/pull/2788 I think we should not hold the release for this as it is not a general use case. We can fix it in the next z-stream release.
I've tested the scenario described in the BZ on a cluster with ODF : 4.12.0-0.nightly-2022-11-25-185455 The exact steps were: Steps to Reproduce: 1. Create a pvc of 2Gi, expand it to 4Gi 2. Create a volume snapshot of the 4Gi pvc 3. Verify that the snapshot "READY TO USE" state is "true" 4. Expand parent pvc to 6GiB 5. Restore snapshot to a pvc of 4Gi And it looks that all the steps were completed successfully. The above event refers to the restore action of PVC of size 4GB. Generated from openshift-storage.rbd.csi.ceph.com_csi-rbdplugin-provisioner-6f5c888c64-nj7kc_037e6ead-1cd9-41f6-84fa-70496a2ca623 Successfully provisioned volume pvc-7aaf4c00-1378-48c0-b715-27bcdc9884b3 Is it possible that something else cause Restore action failure?
Verified in ODF : 4.12.0-0.nightly-2022-11-25-185455 ( see comment#8) .