Summary of resolution: upstream docs will be updated 
Pull Request Made PR https://github.com/vmware-tanzu/velero/pull/4729

Description of problem:
Failed/PartiallyFailed backups hang on Kubernetes although removed from the bucket.
Successfully completed buckups are deleted immediately as expected (see “Object storage sync” in

https://velero.io/docs/v1.7/how-velero-works/).

Checked with both CSI and Restic backups.

Example output:

http://pastebin.test.redhat.com/1018311

Version-Release number of selected component (if applicable): 0.5.3

How reproducible:

Always

Steps to reproduce:

1. Create a successful backup (Completed status)
2. Empty the bucket

• Notice that after few seconds the backup CR is deleted from the cluster as expected

1. Create a backup that will fail or partially fail. To accomplish that, create a stateful app which uses csi driver as storageclass and enable csi as a default plugin). 

1. Empty the bucket

• Notice that the PartiallyFailed backup CR persists

Expected results:

From https://velero.io/docs/v1.7/how-velero-works/, "Object storage sync" section:
"if a backup object exists in Kubernetes but not in object storage, it will be deleted from Kubernetes since the backup tarball no longer exists.”