A proposal to force-drop replication slots to make disabling async

Debezium stops streaming: ERROR: cannot advance replication slot error_ replication slot is active for pid Monitoring Replication Slots Each BDR node keeps: One replication slot per active BDR peer; One group replication slot For example: Toggle Wrap Copy bdrdb error slot The logical replication slot “mylogslot” is also active and permanent It's client has received up to 0

error_ replication slot is active for pid Previous PostgreSQL logs shows usage of the replication slot with 2 different PIDs with a delay between the 2 calls of less than a minute