Horton

Re: add log messages when replication slots become active and inactive (was Re: Is it worth adding ReplicationSlot active_pid to ReplicationSlotPersistentData?)

Peter Smith <smithpb2250@gmail.com>

View thread View on postgr.es
From: Peter Smith <smithpb2250@gmail.com>
To: Bharath Rupireddy <bharath.rupireddyforpostgres@gmail.com>
Cc: Amit Kapila <amit.kapila16@gmail.com>, Alvaro Herrera <alvherre@alvh.no-ip.org>, Greg Stark <stark@mit.edu>, Euler Taveira <euler@eulerto.com>, Kyotaro Horiguchi <horikyota.ntt@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2023-11-07T01:33:07Z
Lists: pgsql-hackers

Attachments

Hi,

FWIW, I used a small script to set up the following environment to
help observe the log messages written by this patch.

Node1 = primary; it creates a PUBLICATION
Node2 = physical standby; the publication is replicated here
Node3 = subscriber; it subscribes to the publication now on the standby Node2

~

In the log files you can see:

1. The "physical" message from the walsender of the primary (Node1)
e.g. 2023-11-06 18:59:58.094 AEDT [17091] LOG:  walsender process with
PID 17091 acquired physical replication slot "pg_basebackup_17091"

2. The "logical" message from the pub/sub walsender of the standby (Node2)
e.g. 2023-11-06 19:16:29.053 AEDT [12774] LOG:  walsender process with
PID 12774 acquired logical replication slot "sub1"

~

PSA the test script and logs

======
Kind Regards,
Peter Smith.
Fujitsu Australia

Commits

  1. Log messages for replication slot acquisition and release.