[Bug] Unclean shutdown handling in k8s Endpoints discovery

[andrewazores]

Current Behavior

See cryostat-report log at cryostatio/cryostat-operator#816 (comment) .

When Cryostat is shutting down, it looks like it observes its own shutdown via the Endpoints Informer, then tries to act upon that event. However, since it is already shutting down, the database connection is going down and the EntityManager/Session gets closed.

Expected Behavior

When shutting down, Cryostat should stop listening for such discovery events, and if any come in late they should be ignored.

Steps To Reproduce

No response

Environment

No response

Anything else?

No response