-
Sub-task
-
Resolution: Done
-
Undefined
-
None
-
None
-
None
-
False
-
-
False
-
ToDo
-
-
-
0
-
0
-
Very Likely
-
0
-
None
-
Unset
-
Unknown
Description of problem:
When applying backup policies in the backup cr
backup spec.resourcepolicy.kind is only respected with lower-level string
tracking upstream PR with the fix:
https://github.com/vmware-tanzu/velero/pull/6804
Version-Release number of selected component (if applicable):
How reproducible:
100%
Steps to Reproduce:
1. Deploy velero with csi pluging
2. Deploy an application with a pod mounted to a PVC with csi storageclass x
3. Create a configmap setting action: skip for storage class X in oadp namespace used to deploy the above app, for example:
For example: <yaml-file> version: v1 volumePolicies: - conditions: storageClass: - x action: type: skip $ oc create cm rp1 --from-file <yaml-file> -n openshift-adp
3. Backup your application ns and apply the above resource policy
setting spec.resourcePolicy.kind to ConfigMap (upper C and M)
apiVersion: velero.io/v1
kind: Backup
metadata:
name: b1
labels:
velero.io/storage-location: default
namespace: openshift-adp-test
spec:
resourcePolicy:
kind: ConfigMap
name: rp1
includedNamespaces:
- <app namespace>
storageLocation: example-gcp-velero-1
ttl: 720h0m0s
Actual results:
the PV was backed up
status: completionTimestamp: "2023-08-10T06:25:20Z" csiVolumeSnapshotsAttempted: 1 <----------- csiVolumeSnapshotsCompleted: 1 <----------- expiration: "2023-09-09T06:24:43Z" formatVersion: 1.1.0 phase: Completed startTimestamp: "2023-08-10T06:24:43Z" version: 1
Expected results:
The backup should not attempt to create CSI snapshot during backup
apiVersion: velero.io/v1 kind: Backup metadata: name: b1 namespace: openshift-adp ... spec: ... status: completionTimestamp: "2023-08-10T06:16:33Z" expiration: "2023-09-09T06:16:25Z" formatVersion: 1.1.0 phase: Completed startTimestamp: "2023-08-10T06:16:26Z" version: 1
Additional info: