From: Andres Freund <andres@anarazel.de>
To: Thomas Munro <thomas.munro@gmail.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Michael Paquier <michael@paquier.xyz>
Cc: Thomas Munro <tmunro@postgresql.org>
Cc: pgsql-committers <pgsql-committers@lists.postgresql.org>
Subject: Re: pgsql: Test replay of regression tests, attempt II.
Date: Wed, 19 Jan 2022 21:24:04 -0800
Message-ID: <20220120052404.sonrhq3f3qgplpzj@alap3.anarazel.de> (raw)
In-Reply-To: <CA+hUKG+nHX+NNjm-ig0zWLxeMiivH8omey5Onfhnxzh6g524Cg@mail.gmail.com>
References: <E1n9Io8-0007At-FQ@gemulon.postgresql.org>
<YeUUiruW+mqNNGIG@paquier.xyz>
<20220117203746.oj43254j5jurbneu@alap3.anarazel.de>
<CA+hUKGLT2UL5_JhmBzUgkdyKfc=5J-gJSQJLysMs4rqLUKLAzw@mail.gmail.com>
<CA+hUKGJDrokTTT3fAEESHiW5TmwxkkJGqe8bR-VFjpg4+hK+fQ@mail.gmail.com>
<374643.1642536944@sss.pgh.pa.us>
<20220118204012.qpfas4fmvc55g4ei@alap3.anarazel.de>
<415981.1642544346@sss.pgh.pa.us>
<20220118230838.amvazn2ig4cf56oy@alap3.anarazel.de>
<CA+hUKG+nHX+NNjm-ig0zWLxeMiivH8omey5Onfhnxzh6g524Cg@mail.gmail.com>
Hi,
On 2022-01-20 17:23:30 +1300, Thomas Munro wrote:
> Having failed to reproduce this locally, I clicked on "re-run tests"
> all afternoon on CI until eventually I captured a failure log[1]
> there, with the smoking gun:
Phew, finally.
> Since this page doesn't require wraparound vacuuming, if we fail to
> conditionally acquire the cleanup lock, this block skips the page.
The most likely source for such a pin is bgwriter, I think?
It's not actually *immediately* obviously why the page doesn't require
wraparound, as VACUUM FREEZE should use a very aggressive one. The reason is
presumably that there are concurrent backends holding back the horizon. Note
that the oldest xmin in the log below (15356 for both vacuums) is older than
the xids from the intentionally failed insert (15465).
I wonder if the easiest way to make this test reliable would be to make the
table a temporary one? That now uses very aggressive horizons, there's no
bgwriter that could pin the page, etc.
Regards,
Andres
2022-01-20 03:58:31.966 UTC [32584][client backend] [pg_regress/reloptions][4/5416:0] LOG: statement: INSERT INTO reloptions_test VALUES (1, NULL), (NULL, NULL);
2022-01-20 03:58:31.967 UTC [32584][client backend] [pg_regress/reloptions][4/5416:15465] ERROR: null value in column "i" of relation "reloptions_test" violates not-null constraint
2022-01-20 03:58:31.967 UTC [32584][client backend] [pg_regress/reloptions][4/5416:15465] DETAIL: Failing row contains (null, null).
2022-01-20 03:58:31.967 UTC [32584][client backend] [pg_regress/reloptions][4/5416:15465] STATEMENT: INSERT INTO reloptions_test VALUES (1, NULL), (NULL, NULL);
2022-01-20 03:58:31.970 UTC [32584][client backend] [pg_regress/reloptions][4/5417:0] LOG: statement: VACUUM (FREEZE, DISABLE_PAGE_SKIPPING) reloptions_test;
2022-01-20 03:58:31.973 UTC [32584][client backend] [pg_regress/reloptions][4/5418:0] LOG: automatic aggressive vacuum of table "regression.public.reloptions_test": index scans: 0
pages: 0 removed, 1 remain, 1 skipped due to pins, 0 skipped frozen
tuples: 0 removed, 0 remain, 0 are dead but not yet removable, oldest xmin: 15356
index scan not needed: 0 pages from table (0.00% of total) had 0 dead item identifiers removed
avg read rate: 372.024 MB/s, avg write rate: 0.000 MB/s
buffer usage: 3 hits, 6 misses, 0 dirtied
WAL usage: 1 records, 0 full page images, 237 bytes
system usage: CPU: user: 0.00 s, system: 0.00 s, elapsed: 0.00 s
2022-01-20 03:58:31.973 UTC [32584][client backend] [pg_regress/reloptions][4/5418:0] STATEMENT: VACUUM (FREEZE, DISABLE_PAGE_SKIPPING) reloptions_test;
2022-01-20 03:58:31.973 UTC [32584][client backend] [pg_regress/reloptions][4/5419:0] LOG: automatic aggressive vacuum of table "regression.pg_toast.pg_toast_35786": index scans: 0
pages: 0 removed, 0 remain, 0 skipped due to pins, 0 skipped frozen
tuples: 0 removed, 0 remain, 0 are dead but not yet removable, oldest xmin: 15356
avg read rate: 790.028 MB/s, avg write rate: 0.000 MB/s
buffer usage: 13 hits, 9 misses, 0 dirtied
WAL usage: 0 records, 0 full page images, 0 bytes
system usage: CPU: user: 0.00 s, system: 0.00 s, elapsed: 0.00 s
2022-01-20 03:58:31.973 UTC [32584][client backend] [pg_regress/reloptions][4/5419:0] STATEMENT: VACUUM (FREEZE, DISABLE_PAGE_SKIPPING) reloptions_test;
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients from your mail client; the link fills in
the headers below: reply via email
Some clients, webmail especially, drop In-Reply-To and References from
a link, and the reply then starts a new thread: use the mbox below.
To: pgsql-committers@postgresql.org
Cc: andres@anarazel.de, thomas.munro@gmail.com, tgl@sss.pgh.pa.us, michael@paquier.xyz, tmunro@postgresql.org
Subject: Re: pgsql: Test replay of regression tests, attempt II.
In-Reply-To: <20220120052404.sonrhq3f3qgplpzj@alap3.anarazel.de>
References: <E1n9Io8-0007At-FQ@gemulon.postgresql.org> <YeUUiruW+mqNNGIG@paquier.xyz> <20220117203746.oj43254j5jurbneu@alap3.anarazel.de> <CA+hUKGLT2UL5_JhmBzUgkdyKfc=5J-gJSQJLysMs4rqLUKLAzw@mail.gmail.com> <CA+hUKGJDrokTTT3fAEESHiW5TmwxkkJGqe8bR-VFjpg4+hK+fQ@mail.gmail.com> <374643.1642536944@sss.pgh.pa.us> <20220118204012.qpfas4fmvc55g4ei@alap3.anarazel.de> <415981.1642544346@sss.pgh.pa.us> <20220118230838.amvazn2ig4cf56oy@alap3.anarazel.de> <CA+hUKG+nHX+NNjm-ig0zWLxeMiivH8omey5Onfhnxzh6g524Cg@mail.gmail.com> <20220120052404.sonrhq3f3qgplpzj@alap3.anarazel.de>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox