[issue] Should we allow to restore from backups that are PartiallyFailed ?

[mpryc]

Should we allow to discover & restore from backups that were not fully success ?

$ velero describe backup mybackup --details
Name:         mybackup
Namespace:    openshift-adp
Labels:       velero.io/storage-location=ts-dpa-1
Annotations:  velero.io/resource-timeout=10m0s
              velero.io/source-cluster-k8s-gitversion=v1.33.5
              velero.io/source-cluster-k8s-major-version=1
              velero.io/source-cluster-k8s-minor-version=33

Phase:  PartiallyFailed (run `velero backup logs mybackup` for more information)


Errors:
  Velero:    message: /Error executing hook error: /command terminated with exit code 1
             message: /Error executing hook error: /command terminated with exit code 1
  Cluster:    <none>
  Namespaces: <none>

[shubham-pampattiwar]

We only care about the data to be restored in cluster on PV/PVC, if those are bound and are browsable by the VM admin/user then we should be good, so yes we can. IMO if we do allow this, we should at least add a warning/condition on the VMFR status somewhere.

[mpryc]

ATM it's not required, because we expect the restoration to succeed with our changes to the relevant plugins and it would be not ideal if we restore from just partial restored PVCs.

We can add this kind of functionality if needed with possible discover part letting user why it did not succeed before proceeding.

Currently the user have all the logs and VMBD object, so it's easy to filter out on the backups that failed to restore fully.