Content-Type: multipart/mixed;
 boundary=a10abe3a9b71d501f39967ebb75d1dba43571e753d9b1cf57a95d87e101a
Mime-Version: 1.0
Date: Tue, 09 Jun 2026 18:06:09 -0300
Message-Id: <DJ4TOYZGE7M8.1U3BNIRMQVXQX@gmail.com>
Subject: Re: Autoprewarm workers terminated due to a segmentation fault
From: "Matheus Alcantara" <matheusssilv97@gmail.com>
To: "Glauber Batista" <glauberrbatista@gmail.com>,
 <pgsql-bugs@lists.postgresql.org>
References: 
 <CAO+_mTQgQyTYwDh=U8iTnsDmOGyWsZJjUV31SmEYwmw6_xY6Bw@mail.gmail.com>
In-Reply-To: 
 <CAO+_mTQgQyTYwDh=U8iTnsDmOGyWsZJjUV31SmEYwmw6_xY6Bw@mail.gmail.com>
Archived-At: 
 <https://www.postgresql.org/message-id/DJ4TOYZGE7M8.1U3BNIRMQVXQX%40gmail.com>
Precedence: bulk

--a10abe3a9b71d501f39967ebb75d1dba43571e753d9b1cf57a95d87e101a
Content-Type: multipart/alternative;
 boundary=9a41f06ce25d6b3d60c1164a9575139f2a7e6b9225c1f41c3b431fa53503

--9a41f06ce25d6b3d60c1164a9575139f2a7e6b9225c1f41c3b431fa53503
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Content-Type: text/plain; charset=UTF-8

Hi,

On Tue Jun 9, 2026 at 3:37 PM -03, Glauber Batista wrote:
> I have an issue with the autoprewarm workers segfaulting during the servi=
ce
> restart. Sometimes, it successfully restarts after a few tries, but usual=
ly
> I need to remove the autoprewarm.blocks file. My setup consists of a
> primary server with two replicas and all of them present the same issue. =
I
> have been using this setup for several years with no issues, but since I
> upgraded to Postgres 18 I'm having it. This is a production database.
>
> Details:
>
> [ ... ]=20
>
> All that said, it seems there's a missing guard-clause at line 649. I
> didn't spend much time reading the code, but it's clearly accessing a
> position in the array that is not allocated.
>

Thank you for the report!

I've managed to reproduce the issue with the following:

create table test_warm (id int, data text);
-- insert enough rows to generate too many pages on an index
insert into test_warm select g, repeat('a', 100) from generate_series(1, 50=
00000) g;
create index warm_idx on test_warm(id);

-- force read the index entirely into shared_buffers
select count(*) from test_warm where id > 0;

Then pg_ctl stop and pg_ctl start will start failing with the following log=
s:

2026-06-09 17:47:40.924 -03 [23025] LOG:  shutting down
2026-06-09 17:47:40.925 -03 [23025] LOG:  checkpoint starting: shutdown fas=
t
2026-06-09 17:47:41.033 -03 [23022] LOG:  database system is shut down
2026-06-09 17:47:49.830 -03 [23172] LOG:  starting PostgreSQL 19beta1 on aa=
rch64-darwin, compiled by clang-17.0.0, 64-bit
2026-06-09 17:47:49.842 -03 [23172] LOG:  database system is ready to accep=
t connections
2026-06-09 17:47:49.917 -03 [23172] LOG:  background worker "autoprewarm wo=
rker" (PID 23182) was terminated by signal 11: Segmentation fault: 11
2026-06-09 17:47:49.917 -03 [23172] LOG:  terminating any other active serv=
er processes
2026-06-09 17:47:49.918 -03 [23172] LOG:  all server processes terminated; =
reinitializing

Wondering if the following would be enough?

  /* Advance i past all the blocks just prewarmed. */
  i =3D p.pos;
+ if (i >=3D apw_state->prewarm_stop_idx)
+         break;
+
  blk =3D block_info[i];


With this change on a fresh start I got very similar shared hit buffers
on the first and second execution, so I think that the
autoprewarm.blocks was reloaded successfully?

postgres=3D# explain(analyze, costs off, timing off, summary off) select co=
unt(*) from test_warm where id > 0;
                                    QUERY PLAN
---------------------------------------------------------------------------=
--------
 Finalize Aggregate (actual rows=3D1.00 loops=3D1)
   Buffers: shared hit=3D15965 read=3D70242
   ->  Gather (actual rows=3D3.00 loops=3D1)
         Workers Planned: 2
         Workers Launched: 2
         Buffers: shared hit=3D15965 read=3D70242
         ->  Partial Aggregate (actual rows=3D1.00 loops=3D3)
               Buffers: shared hit=3D15965 read=3D70242
               ->  Parallel Seq Scan on test_warm (actual rows=3D1666666.67=
 loops=3D3)
                     Filter: (id > 0)
                     Buffers: shared hit=3D15965 read=3D70242

postgres=3D# explain(analyze, costs off, timing off, summary off) SELECT co=
unt(*) FROM test_warm WHERE id > 0;
                                    QUERY PLAN
---------------------------------------------------------------------------=
--------
 Finalize Aggregate (actual rows=3D1.00 loops=3D1)
   Buffers: shared hit=3D15970 read=3D70237
   ->  Gather (actual rows=3D3.00 loops=3D1)
         Workers Planned: 2
         Workers Launched: 2
         Buffers: shared hit=3D15970 read=3D70237
         ->  Partial Aggregate (actual rows=3D1.00 loops=3D3)
               Buffers: shared hit=3D15970 read=3D70237
               ->  Parallel Seq Scan on test_warm (actual rows=3D1666666.67=
 loops=3D3)
                     Filter: (id > 0)
                     Buffers: shared hit=3D15970 read=3D70237


Thoughts?

--
Matheus Alcantara
EDB: https://www.enterprisedb.com

--9a41f06ce25d6b3d60c1164a9575139f2a7e6b9225c1f41c3b431fa53503--

--a10abe3a9b71d501f39967ebb75d1dba43571e753d9b1cf57a95d87e101a
Content-Transfer-Encoding: base64
Content-Disposition: attachment;
 filename=0001-Fix-out-of-bounds-access-in-autoprewarm-worker.patch
Content-Type: text/plain; charset=utf-8;
 name=0001-Fix-out-of-bounds-access-in-autoprewarm-worker.patch
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--a10abe3a9b71d501f39967ebb75d1dba43571e753d9b1cf57a95d87e101a--