Re: Remove wal_[sync|write][_time] from pg_stat_wal and track_wal_io_timing

From: Bertrand Drouvot <bertranddrouvot(dot)pg(at)gmail(dot)com>
To: Michael Paquier <michael(at)paquier(dot)xyz>
Cc: pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Re: Remove wal_[sync|write][_time] from pg_stat_wal and track_wal_io_timing
Date: 2025-02-20 08:11:12
Message-ID: Z7bjoBhQqgDmo+TF@ip-10-97-1-34.eu-west-3.compute.internal
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi,

On Thu, Feb 20, 2025 at 02:37:18PM +0900, Michael Paquier wrote:
> On Wed, Feb 19, 2025 at 09:24:41AM +0000, Bertrand Drouvot wrote:
> > That's right but that would mean almost duplicating the pg_stat_wal related
> > stuff (because it can't be removed from the doc until the fields are gone). I
> > think it's simpler to do it as proposed initially (the end result is the same).
>
> After sleeping on it, I still saw no reason to not apply the changes
> from 0002 in wal.sgml to describe that the stats for the writes/fsyncs
> are in pg_stat_io rather than pg_stat_wal for the "WAL configuration"
> section, so done that.

I see, I did not get that you wanted to get rid of the pg_stat_wal part before
removing the fields. Makes sense that way to "just" replace the pg_stat_wal
by pg_stat_io first in the doc as you've done in 4538bd3f1dd.

> and I've moved that as a paragraph not under the tuning part, to make
> it a more general statement with its link to "WAL configuration",
> which is a very popular link for pg_stat_io.

Makes more sense, agree.

> Attached is the rest, as of v3-0002.

LGTM.

Regards,

--
Bertrand Drouvot
PostgreSQL Contributors Team
RDS Open Source Databases
Amazon Web Services: https://aws.amazon.com

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Bertrand Drouvot 2025-02-20 08:32:47 Re: Draft for basic NUMA observability
Previous Message Michael Paquier 2025-02-20 08:02:42 Re: Add Pipelining support in psql