So we have a configuration where we connect to a cluster with primary, secondary replicas. There is a case which we can't reproduce where rethink-migrate continues to run indefinitely without exiting, partly because not all the replicas are ready.
1) If only the primary replica is ready_for_writes rethink migrate should run. Does it?
2) If rethink continues to run indefinitely, can we have a timeout, per migration?
So we have a configuration where we connect to a cluster with primary, secondary replicas. There is a case which we can't reproduce where rethink-migrate continues to run indefinitely without exiting, partly because not all the replicas are ready.
1) If only the primary replica is ready_for_writes rethink migrate should run. Does it? 2) If rethink continues to run indefinitely, can we have a timeout, per migration?