Description of problem:

While re-testing the https://issues.redhat.com/browse/RHODS-1997 I found that the operator re-deployment is failing.

Once the re-deployment is triggered, a new rhods-operator pod is instantiated and it gets continuously restarted, hitting the CrashLoopBackOff error. 

The latest RHODS operator condition says: "installing: waiting for deployment rhods-operator to become ready: deployment "rhods-operator" waiting for 1 outdated replica(s) to be terminated". The process doesn't stop, and the rhod operators remains in "Installation". It will probably fail when it'll reach the timeout.

It is the first time I  encounter this situation, so i guess it might be a regression due to the fix applied in https://issues.redhat.com/browse/RHODS-1949 . 

Prerequisites (if any, like setup, operators/versions):

Install RHODS v1.1.1-52

Steps to Reproduce

1. trigger the RHODS operator re-deployment
2. check the operator pod

Actual results:

RHODS operator re-deployment is failing

Expected results:

RHODS operator re-deployment is successful 

Reproducibility (Always/Intermittent/Only Once):

not sure, i tried only once

Build Details:

RHODS v1.1.1-52

Workaround:

Additional info:

I attached some screenshots to show the situation and the operator pod's logs