The old sdk leader election always assumed the lock could be found in the same namespace as the pod. this is an issue if the operator is running in a different cluster than the apiserver its talking to, such as in a hypershift topology
*** This bug has been marked as a duplicate of bug 1965113 ***