Re: [PATCH] Expose checkpoint reason to completion log messages.

Michael Banck <mbanck@gmx.net>

From: Michael Banck <mbanck@gmx.net>
To: Fujii Masao <masao.fujii@gmail.com>
Cc: Soumya S Murali <soumyamurali.work@gmail.com>, Andres Freund <andres@anarazel.de>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>, Álvaro Herrera <alvherre@kurilemu.de>, melanieplageman@gmail.com, juanjo.santamaria@gmail.com
Date: 2026-01-06T10:38:47Z
Lists: pgsql-hackers
Hi,

On Tue, Jan 06, 2026 at 02:41:16PM +0900, Fujii Masao wrote:
> On Thu, Jan 1, 2026 at 5:09 PM Soumya S Murali
> <soumyamurali.work@gmail.com> wrote:
> > On Thu, Dec 18, 2025 at 5:49 AM Andres Freund <andres@anarazel.de> wrote:
> > > On 2025-12-01 16:48:56 +0530, Soumya S Murali wrote:
> > > > This patch is an update after reworking the “checkpoint reason” changes as
> > > > a standalone patch, separate from the pg_stat_checkpointer additions as
> > > > suggested [1].
> 
> This kind of information already seems to be included in the
> checkpoint starting log message, for example:
> 
>     LOG:  checkpoint starting: fast force wait
> 
> Why do you want to include essentially the same information in the
> checkpoint ending log message as well?

I do think it is useful; the checkpoint finished message has a lot more
information than the checkpoint starting meessage (which basically only
mentions the reason) and when you extract log messages you currently
need to backtrack from the checkpoint finished message to the
corresponding checkpoint starting message to find out the type. Seeing
how the checkpoint finished message is so much more verbose anyway, I
think adding the reason to it is reasonably and helpful.


Michael



Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →
  1. Log checkpoint request flags in checkpoint completion messages.