Horton

Re: Report oldest xmin source when autovacuum cannot remove tuples

Shinya Kato <shinya11.kato@gmail.com>

View thread
From: Shinya Kato <shinya11.kato@gmail.com>
To: Japin Li <japinli@hotmail.com>
Cc: wenhui qiu <qiuwenhuifx@gmail.com>, Sami Imseih <samimseih@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2026-05-23T05:22:48Z
Lists: pgsql-hackers

Attachments

On Sat, May 23, 2026 at 10:40 AM Japin Li <japinli@hotmail.com> wrote:
> Thanks for updating the patch. LGTM. Just one nitpick.
>
> +       int                     pid;                    /* backend pid (0 for slots) */
>
> Per the code, the prepared transaction is also associated with a PID of zero.

Thanks for the review! You're right, prepared transactions also have
pid 0. Fixed the comment to mention that.

Additionally, I forgot to update meson.build, which caused the tests
to fail. I have fixed that in the attached patch.

-- 
Best regards,
Shinya Kato
NTT OSS Center