MIME-Version: 1.0
References: <0dce35ca-24fa-4234-99fe-071bd5725a32@ya.ru>
 <c6b71dd5f6839962adef8b1481d9de348ccce661.camel@cybertec.at>
 <aeec68b8-9602-4572-b530-2c09a892df03@ya.ru>
 <04dc073cd84313cb9b711833f905a07d099a5f28.camel@cybertec.at>
 <99731769513229@mail.yandex.ru>
In-Reply-To: <99731769513229@mail.yandex.ru>
From: Kristjan Mustkivi <sonicmonkey@gmail.com>
Date: Tue, 27 Jan 2026 14:10:07 +0200
Message-ID: 
 <CAOQPKavjTPzfrfvyR2DM4ALo7ea7vrbc3rh2h-qB_2S0C7wijw@mail.gmail.com>
Subject: Re: autoanalyze did not run
To: Pgsql-admin <pgsql-admin@lists.postgresql.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Archived-At: 
 <https://www.postgresql.org/message-id/CAOQPKavjTPzfrfvyR2DM4ALo7ea7vrbc3rh2h-qB_2S0C7wijw%40mail.gmail.com>
Precedence: bulk

On Tue, Jan 27, 2026 at 1:40=E2=80=AFPM =D0=9E=D0=BB=D0=B5=D0=B3 =D0=A1=D0=
=B0=D0=BC=D0=BE=D0=B9=D0=BB=D0=BE=D0=B2 <splarv@ya.ru> wrote:
> ----------------
> =D0=9A=D0=BE=D0=BC=D1=83: =D0=9E=D0=BB=D0=B5=D0=B3 =D0=A1=D0=B0=D0=BC=D0=
=BE=D0=B9=D0=BB=D0=BE=D0=B2 (splarv@ya.ru), Pgsql-admin (pgsql-admin@lists.=
postgresql.org);
> =D0=A2=D0=B5=D0=BC=D0=B0: autoanalyze did not run;
> 26.01.2026, 19:03, "Laurenz Albe" <laurenz.albe@cybertec.at>:
>
> On Mon, 2026-01-26 at 17:35 +0300, =D0=9E=D0=BB=D0=B5=D0=B3 =D0=A1=D0=B0=
=D0=BC=D0=BE=D0=B9=D0=BB=D0=BE=D0=B2 wrote:
>
>  > > I try to investigate why autoanalyze did not run in time of our init=
ial
>  > > loading data. Yes, I know, running analyze manually is highly
>  > > recommended in such case. But is must run automatically too.
>  > >
>  > > PostgreSQL 17.7 (Debian 17.7-3.pgdg13+1) on x86_64-pc-linux-gnu,
>  > > compiled by gcc (Debian 14.2.0-19) 14.2.0, 64-bit (from PGDG)
>  >
>  > Are you sure that you committed the transaction?
>
>  I am very sure, thats why there are 13844347 rows in that table. By sing=
le copy.
>  >
>  > Is "track_activities" set to "on"?
>
>  I checked, yep.
>
>
> Then the only explanation is that all autovacuum workers are active, and =
this
> table is starved. See how many autovacuum workers are visible in pg_stat_=
activity
> and compare that to autovacuum_max_workers.
>
> Yours,
> Laurenz Albe
>
> Now 0. But they worked. The reason is other. Look, first of all, there wa=
s not a statistic reset.
> SELECT stats_reset is null FROM pg_stat_database WHERE datname =3D curren=
t_database();
>  ?column?\c
> ----------
>  t
> (1 row)
>  select analyze_count,autoanalyze_count,reltuples, n_live_tup, n_tup_ins,=
 n_tup_upd, n_mod_since_analyze from pg_stat_user_tables as s join pg_class=
 as c on (s.relid=3Dc.oid) where relkind=3D'r' and reltuples>0 limit 10;
>  analyze_count | autoanalyze_count |   reltuples   | n_live_tup | n_tup_i=
ns | n_tup_upd | n_mod_since_analyze
> ---------------+-------------------+---------------+------------+--------=
---+-----------+---------------------
>              0 |                 0 |             8 |          0 |        =
 0 |         0 |                   0
>              0 |                 0 |         76596 |          0 |        =
 0 |         0 |                   0
>              0 |                 0 |        140997 |          0 |        =
 0 |         0 |                   0
>              0 |                 0 |  2.088436e+07 |          0 |        =
 0 |         0 |                   0
>              0 |                 0 | 1.2661012e+07 |          0 |        =
 0 |         0 |                   0
>              0 |                 0 |  2.288401e+07 |          0 |        =
 0 |         0 |                   0
>              0 |                 0 |         99926 |          0 |        =
 0 |         0 |                   0
>              0 |                 0 | 1.5620866e+07 |          0 |        =
 0 |         0 |                   0
>              0 |                 0 |         17511 |          0 |        =
 0 |         0 |                   0
>              0 |                 0 |        400926 |          0 |        =
 0 |         0 |                   0
> The reason is the n_mod_since_analyze is 0. Thats why autoanalyze didn't =
started.

Hello,

I do apologize if I mix something up now, but I would expect COPY to
have an effect on "n_ins_since_vacuum" but not "n_mod_since_analyze".
COPY inserts rows, it does not update them.

With best regards,
--=20
Kristjan Mustkivi

Email: kristjan.mustkivi@gmail.com