Something that has worked for me was to clear the volumes of the mongo cluster and reconfigure the replica set. I had stale configuration there and that is why I had issue.