[BUG]: When VolumeGroup Snapshot ReadyToUse, the VolumeSnapshot don't have Status about csm HOT 9 CLOSED

please retest with VG image 0.3.0 , seems related to issue reported in #67

from csm.

I still saw this issue in 0.3.0 VolumeGroup Snapshotter.

from csm.

Hi @phuongatemc -- can you give us access to the cluster with the issue? Also, could you give us:

1. kubectl describe volumesnapshot xxx -n namespace output where status is missing
2. kubectl logs of the external-snapshotter pod?
3. The output of kubectl logs controller vg-snapshotter -n vxflexos
4. The version of external-snapshotter you are running
5. Please run the verify.sh script located in the scripts folder and send us the output.

from csm.

I think this may be a timing issue because if I created the VolumeGroup Snapshot manually using kubectl commands, I've never seen this issue. I only saw this issue when I used golang code to fetch the VolumeSnapshots after they are created in the VolumeGroup Snapshot (Status.ReadyToUse == true). And even with golang code, it is intermittent.

from csm.

talked to customer , suggest wait for volumesnapshot status to be updated , it may take a little time because external-snapshotter is responsible for updating status . if this is indeed the fix please confirm

from csm.

I have implemented in my controller code to validate VolumeSnapshot and VolumeSnapshotContent and retry if they are not ready as discussed with Jai Kumar. So far I couldn't see the problem again.

from csm.

created story action item to fix this in vgs controller.

from csm.

Marking this as closed per the above comment.

from csm.

This issue was fixed as part of dell/csi-volumegroup-snapshotter#14.

from csm.