Date: Tue, 21 May 2024 21:20:36 +0200 (CEST)
From: Dimitrios Apostolou <jimis@gmx.net>
To: Kashif Zeeshan <kashi.zeeshan@gmail.com>
cc: Muhammad Imtiaz <imtiaz.m@bitnine.net>, pgsql-general@lists.postgresql.org
Subject: Re: pg_stat_io clarifications: background worker, writes and reads
In-Reply-To: <CAAPsdhd4tpC6jHmg0ED4rvBZcLTf3Qx+pgPxC3XTjtRZTbAjTQ@mail.gmail.com>
Message-ID: <89bc1c8f-aac1-926c-38bb-1bbf55bec4b0@gmx.net>
References: <2e9b2b7b-79a1-a7f2-e05e-9cefda601c09@gmx.net> <CANwR1OSB93AwAqNfwheCq6xA5KWfpJqh3BkuU0Ta-wPzVzZUYg@mail.gmail.com> <12733373-8f3c-550e-771d-fa7b5ff9343d@gmx.net> <CAAPsdhd4tpC6jHmg0ED4rvBZcLTf3Qx+pgPxC3XTjtRZTbAjTQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="0-1747678927-1716319237=:5296"
UI-OutboundReport: notjunk:1;M01:P0:uvKM6CrAEo0=;Y4jCGY6WUcjwjZ0ab9MbxRsKPAW
 PEwgox93srbTHgr5WJabzbHO0twCfpZeSusaD1G3vngQTbUcWVSzLZ2mJUkI60uZeBuNRoNZO
 N40QLr85Svwx1yUgLA5wMWWRDk7WjcC8ZU0LJ4j67NsuczlPgMA533xl8fz2ERajvMIXLHgKV
 wlfx9FxlZkj8THz8Gj0W1A2RFWy4Jq47BB+qMUbBjynZSEHwRTXcyv4iEc1owUbFOBSqqDb3X
 ZuTMOHYrbRimK50e3M/0fQ92RskyqJAEISzfmEIgyhjX/qQbOG9bmbKAfa6RKd9G32veLSuIB
 0l1p0wG0QU9x3kwdudNv/Stb0IKJ/FjKMTaNWE2a6RNlXv4O/iM2hGrKcFsuxWvGHl3PKDILt
 KcxI6wIPHZXgu845vN1P7WrB1JPIYyWThHO3pEH1lFbvMpF1rIS0CsdwJ7JVvall+mAUsb+Va
 KuBwUb12Gg7jKySKAKuFB8tLIetNEHBsCuJesVJkagW7P7TlkvHB0PUiNArVHn3/rCxuRbCvH
 Gok0u+CVO7GBEU15rGZ3624SQ44GTBXHtE3XHtFsI+CFULCxuuU1JoCE05oNZvzqo1tJzd8bk
 HEiIG5EmXs9duQQXJGw1Mlv91e08JngFHwV2dOwu51LQEcqhv8T2ipoXE1uyu5Ft9byQMI4eA
 OWmBGpVKQjkWPxsyhuVbMVa7N9Lg06pwVP0J1pwAIU2/M0zckkSeZcoleF9rj+ctaNQywEsZJ
 8l+ISFkiuMRKLVHIyI7a4fHxp1KNOQMrICW2esd2eQamikFh4F5ALNC8bROpA4nyxP9n1GNUH
 ZI0KkxTyyDKQxcx3IGcuXV+2YyyT3oVr8x6s6YynjjS/A=
Archived-At: <https://www.postgresql.org/message-id/89bc1c8f-aac1-926c-38bb-1bbf55bec4b0%40gmx.net>
Precedence: bulk

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.

--0-1747678927-1716319237=:5296
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: quoted-printable

Hi Kashif, just to clarify my question, are you saying that the I/O from
parallel workers is recorded into the "background worker" entry of
pg_stat_io?

Thanks,
Dimitris


On Wed, 15 May 2024, Kashif Zeeshan wrote:

> Hi
> parallel worker are used for parallel execution of the queries and you c=
an find the help in the below link.
>
> https://www.postgresql.org/docs/current/how-parallel-query-works.html
>
> Its controlled by following parameters.
>
> max_worker_processes =3D 6
> max_parallel_workers_per_gather =3D 3
> max_parallel_workers =3D 6
> The limit of concurrent parallel workers for the whole cluster is max_pa=
rallel_workers, which must be =E2=89=A4 max_worker_processes. The limit of=
 parallel workers per query is max_parallel_workers_per_gather.
>
> Thanks
> Kashif Zeeshan
> Bitnine Global
>
> On Wed, May 15, 2024 at 5:59=E2=80=AFPM Dimitrios Apostolou <jimis@gmx.n=
et> wrote:
>       So what is this particular "background worker" I'm seeing, given t=
hat I
>       have no replication or extensions?
>
>       Searching the logs I found entries like the following:
>
>       LOG:=C2=A0 background worker "parallel worker" (PID 93384) exited =
with exit code 1
>
>       This got logged when I killed a simple SELECT query that took too =
long
>       doing parallel seqscans. Could it be that the entry in pg_stat_io =
named
>       "background worker" also includes the parallel workers from a SELE=
CT
>       query?
>
>       Thanks,
>       Dimitris

--0-1747678927-1716319237=:5296--