Re: Intermittent buildfarm failures on wrasse
Andres Freund <andres@anarazel.de>
From: Andres Freund <andres@anarazel.de>
To: pgsql-hackers@lists.postgresql.org, Tom Lane <tgl@sss.pgh.pa.us>
Cc: Noah Misch <noah@leadboat.com>, Peter Geoghegan <pg@bowt.ie>,
David Rowley <dgrowleyml@gmail.com>,
PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-04-15T15:19:54Z
Lists: pgsql-hackers
Hi, On April 15, 2022 11:12:10 AM EDT, Andres Freund <andres@anarazel.de> wrote: >Hi, > >On 2022-04-15 10:15:32 -0400, Tom Lane wrote: >> The morning's first result is that during a failing run, >> the vacuum in test_setup sees >> >> 2022-04-15 16:01:43.064 CEST [4436:75] pg_regress/test_setup LOG: statement: VACUUM ANALYZE tenk1; >> 2022-04-15 16:01:43.064 CEST [4436:76] pg_regress/test_setup LOG: vacuuming "regression.public.tenk1" >> 2022-04-15 16:01:43.064 CEST [4436:77] pg_regress/test_setup STATEMENT: VACUUM ANALYZE tenk1; >> 2022-04-15 16:01:43.071 CEST [4436:78] pg_regress/test_setup LOG: finished vacuuming "regression.public.tenk1": index scans: 0 >> pages: 0 removed, 345 remain, 345 scanned (100.00% of total) >> tuples: 0 removed, 10000 remain, 0 are dead but not yet removable >> removable cutoff: 724, older by 26 xids when operation ended >> index scan not needed: 0 pages from table (0.00% of total) had 0 dead item identifiers removed >> avg read rate: 2.189 MB/s, avg write rate: 2.189 MB/s >> buffer usage: 695 hits, 2 misses, 2 dirtied >> WAL usage: 1 records, 0 full page images, 188 bytes >> system usage: CPU: user: 0.00 s, system: 0.00 s, elapsed: 0.00 s >> 2022-04-15 16:01:43.071 CEST [4436:79] pg_regress/test_setup STATEMENT: VACUUM ANALYZE tenk1; > >The horizon advancing by 26 xids during tenk1's vacuum seems like quite >a bit, given there's no normal concurrent activity during test_setup. > > >> In fact, right after initdb pg_controldata shows >> Latest checkpoint's NextXID: 0:724 >> Latest checkpoint's oldestXID: 716 > >So that's the xmin that e.g. the autovac launcher ends up with during >start... > >If I make get_database_list() sleep for 5s within the scan, I can >reproduce on x86-64. Off for a bit, but I realized that we likely don't exclude the launcher because it's not database associated... -- Sent from my Android device with K-9 Mail. Please excuse my brevity.
Commits
-
Repurpose PROC_COPYABLE_FLAGS as PROC_XMIN_FLAGS
- 8d9d1286acf9 14.4 landed
- 8d061acd12af 15.0 landed
- 5fd0cccc116b 13.8 landed
-
Tighten ComputeXidHorizons' handling of walsenders.
- 6fea65508a1a 15.0 landed
-
Adjust VACUUM's removable cutoff log message.
- 357c8455e649 15.0 landed
-
Temporarily add some probes of tenk1's relallvisible in create_index.sql.
- 5bb2b6abc8d6 15.0 landed
-
Set synchronous_commit=on in test_setup.sql.
- 02fea8fdda65 15.0 cited
-
Rearrange core regression tests to reduce cross-script dependencies.
- cc50080a828d 15.0 cited
-
snapshot scalability: Don't compute global horizons while building snapshots.
- dc7420c2c927 14.0 cited