Horton

Re: AW: AW: Replication Testing- How to introduce a Lag

Laurenz Albe <laurenz.albe@cybertec.at>

View thread View on postgr.es
From: Laurenz Albe <laurenz.albe@cybertec.at>
To: "Subramanian,Ramachandran" <ramachandran.subramanian@alte-leipziger.de>, "pgsql-novice@lists.postgresql.org" <pgsql-novice@lists.postgresql.org>
Date: 2026-03-21T08:43:10Z
Lists: pgsql-novice 
On Sat, 2026-03-21 at 07:49 +0000, Subramanian,Ramachandran wrote:
> > Look at "pg_stat_replication" on the primary server.
> 
> If I just look at the write_lag columns, it would give me a good idea of the situation --  Is my understanding correct ?

Yes, but if no WAL gets generated and replicated for a while, the columns will
become NULL, because then delay measured in time makes no sense.

Therefore, I prefer to use "replay_lsn" and friends and measure the lag in bytes.

Yours,
Laurenz Albe