Multi-Cluster Replication config changes corrupting database

You are correct. This case is not well handled. This error means the active cluster is still set the IsGlobalNamespaceEnabled as False. But I believe you have changed it to True.

To unblock you, you can clean up the cluster_metadata_info table in active cluster and start the services. Or you can also clean up the DB and initialize the active cluster again.

At the same time, I will make a change to cover this use.