It would be helpful for users to propagate the Replicator failure so that the /status endpoint can report the failure reason rather than having to force users to examine the connector logs themselves to determine the reason for failure.