repmgr keeps sending error: replication slot "repmgr_slot_1" is active for PID 32283

[pham lan]

Hello,

I see a lot of strange errors in postgres log:

2022-03-29 02:10:47.694 UTC [19999] repmgr@[unknown] ERROR:  replication slot "repmgr_slot_1" is active for PID 32283

So i have a cluster of 3 postgres-13 nodes. Whenever i do the switch over the master to the new node, in the new master DB, i got these Error in the log. It appeas every second and fill up the log.

I checked and see that new DB master could stream WAL to the other slave nodes correctly
# ps -ef|grep wal
postgres 28753 37430  0 Mar28 ?        00:01:51 postgres: 13/main: walsender repmgr 10.231.252.33(39320) streaming CF0/6854C7F8
postgres 32283 37430  0 Mar28 ?        00:01:35 postgres: 13/main: walsender repmgr 10.231.252.31(56186) streaming CF0/6854C7F8

postgres=# select * from pg_replication_slots ;
   slot_name   | plugin | slot_type | datoid | database | temporary | active | active_pid | xmin | catalog_xmin | restart_lsn  | confirmed_flush_lsn | wal_status | safe_wal_size
---------------+--------+-----------+--------+----------+-----------+--------+------------+------+--------------+--------------+---------------------+------------+---------------
 repmgr_slot_3 |        | physical  |        |          | f         | t      |      28753 |      |              | CF0/6862FB18 |                     | reserved   |
 repmgr_slot_1 |        | physical  |        |          | f         | t      |      32283 |      |              | CF0/6862FB18 |                     | reserved   |

Does anyone experience the same issue?

Any thought here is appreciated!

Best regards,