ISAR Warning: CephFSStaleSubvolume

[DanielFroehlich]

Cluster is complaining since quite a while about
Number of CephFS subvolumes are greater than the number of PersistentVolumes in the cluster.

Runbook:
https://github.com/openshift/runbooks/blob/master/alerts/openshift-container-storage-operator/CephFSStaleSubvolume.md

[DanielFroehlich]

Oha, somethings wrong:

% odf subvolume ls --stale
Filesystem  Subvolume  SubvolumeGroup  State
Error: Error EAGAIN: subvolume "csi-vol-13384555-e06d-4669-a56d-9e4e3274c89f" is not ready for operation "info"
. failed to run command. command terminated with exit code 11%!(EXTRA string=failed to get subvolume info)
Error: failed to get subvolume state: "csi-vol-13384555-e06d-4669-a56d-9e4e3274c89f" "Error EAGAIN: subvolume \"csi-vol-13384555-e06d-4669-a56d-9e4e3274c89f\" is not ready for operation \"info\"\n. failed to run command. command terminated with exit code 11"

There seems to be no PV with that csi-vol image:

% oc get pv -o jsonpath='{range .items[?(@.spec.csi.volumeAttributes.imageName=="csi-vol-13384555-e06d-4669-a56d-9e4e3274c89f")]}{.metadata.name}{"\n"}{end}'

%

```

[DanielFroehlich]

@rbo I am inclined to delete this subvolume. do you agree?

[rbo]

Yes, it's a lab. Delete it.

[DanielFroehlich]

Gone with the wind:

odf ceph fs subvolume rm ocs-storagecluster-cephfilesystem csi-vol-13384555-e06d-4669-a56d-9e4e3274c89f csi --force

Looks like no stale subvolumes anymore:

% odf subvolume ls --stale                                                                                           
Filesystem  Subvolume  SubvolumeGroup  State
```

Lets see if this clears the alert...

[DanielFroehlich]

...it did! Closing issue