Description of problem:

OADP Operator may fail silently when cloud credentials provide insufficient permissions.

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

1.2.3

How reproducible:

Unknown

Steps to Reproduce:

1. Create a DPA, backupstoragelocation, and anything else needed for OADP to run on a cluster
2. In the DPA, specify a set of credentials which do not permit the operator to list/view S3 buckets in AWS OR remove the required permission after operator pod enters a Running state.
3. Operator pod will continue to report a Running state, examining its logs will reveal that its manager container is actually panicking with each reconcile

Actual results:

OADP remains in a Running state, despite its manager container panicking with each reconcile

Expected results:

OADP should enter a CrashLoopBackOff state when panicking, to alert cluster administrators that the component is failing.

Additional info: