Re: Separate GUC for replication origins
Peter Eisentraut <peter@eisentraut.org>
From: Peter Eisentraut <peter@eisentraut.org>
To: Amit Kapila <amit.kapila16@gmail.com>, Euler Taveira <euler@eulerto.com>
Cc: Masahiko Sawada <sawada.mshk@gmail.com>,
pgsql-hackers@lists.postgresql.org
Date: 2025-03-07T14:47:53Z
Lists: pgsql-hackers
On 07.03.25 04:51, Amit Kapila wrote: >>> I agree that the originally proposed name max_replication_origins is not >>> good, because you can "create" (using pg_replication_origin_create()) >>> more than the configured maximum. What is the term for what the setting >>> actually controls? How many are "active"? "In use"? Per session? etc. >>> >> It controls the number of active sessions using origin. The idea is >> that to track replication progress via replication_origin we need to >> do replorigin_session_setup(). If you look in the code, we have used >> the term replorigin_session* in many places, so we thought of naming >> this as max_replication_origin_sessions. But the other options could >> be max_active_replication_origins or max_replication_origins_in_use. >> >> >> The word "session" is correlated to "replication origin" but requires some >> knowledge to know the replication progress tracking design. The word "active" >> can express the fact that it was setup and is currently in use. I vote for >> max_active_replication_origins. >> > Sounds reasonable. Let's go with max_active_replication_origins then, > unless people think otherwise. Is that maximum active for the whole system, or maximum active per session, or maximum active per created origin, or some combination of these?
Commits
-
Add GUC option to control maximum active replication origins.
- 04ff636cbce4 18.0 landed
-
Better document logical replication parameters
- a8500750ca0a 16.0 cited
-
Introduce replication progress tracking infrastructure.
- 5aa2350426c4 9.5.0 cited