MIME-Version: 1.0
From: Fabrice Chapuis <fabrice636861@gmail.com>
Date: Thu, 28 Sep 2023 13:11:41 +0200
Message-ID: <CAA5-nLCrbc90DHbX_9Yav72hEXUetkZ9iO8zmKvCvYNZkDHx9g@mail.gmail.com>
Subject: wal recycling problem
To: PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Content-Type: multipart/alternative; boundary="000000000000fcd57e0606695f71"
Archived-At: <https://www.postgresql.org/message-id/CAA5-nLCrbc90DHbX_9Yav72hEXUetkZ9iO8zmKvCvYNZkDHx9g%40mail.gmail.com>
Precedence: bulk

--000000000000fcd57e0606695f71
Content-Type: text/plain; charset="UTF-8"

Hello,

I have a question about the automatic removal of unused WAL files. When
loading data with pg_restore (200Gb) we noticed that a lot of WALs files
are generated and they are not purged automatically nor recycled despite
frequent checkpoints, then pg_wal folder (150Gb) fill and become out of
space.
We have a cluster of 2 members (1 primary and 1 standby) with Postgres
version 14.9 and 2 barman server, slots are only configured for barman,
barman is version 3.7.
The archive command is desactivated (archive_command=':')
I use pg_archivecleanup (with the wal file generated from the last
checkpoint in parameter) to remove files manually before the limit of 150Gb
so that the restore can terminate.

Why does postgres do not this cleanup automatically, which part of the code
is responsible for removing or recycling the wals?

Thanks for your help

Fabrice

--000000000000fcd57e0606695f71
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><div>Hello,</div><div><br></div><div>I ha=
ve a question about the automatic removal of unused WAL files. When loading=
 data with pg_restore (200Gb) we noticed that a lot of WALs files are gener=
ated and they are not purged automatically nor recycled despite frequent ch=
eckpoints, then pg_wal folder (150Gb) fill and become out of space.</div><d=
iv>We have a cluster of 2 members (1 primary and 1 standby) with Postgres v=
ersion 14.9 and 2 barman server, slots are only configured for barman, barm=
an is version 3.7.</div><div>The archive command is desactivated (archive_c=
ommand=3D&#39;:&#39;)</div><div>I use pg_archivecleanup (with the wal file =
generated from the last checkpoint in parameter) to remove files manually b=
efore the limit of 150Gb so that the restore can terminate.</div><div><br><=
/div><div>Why does postgres do not this cleanup automatically, which part o=
f the code is responsible for removing or recycling the wals?</div><div><br=
></div><div>Thanks for your help</div><div><br></div><div>Fabrice</div></di=
v></div>

--000000000000fcd57e0606695f71--