Re: Failures with wal_consistency_checking and 13~ - Mailing list pgsql-hackers

From Ranier Vilela
Subject Re: Failures with wal_consistency_checking and 13~
Date
Msg-id CAEudQApwC6Dkd5CbwTyPoi8YeUb9MFTyxYO42a1PY4M=6k8y2Q@mail.gmail.com
Whole thread Raw
In response to Failures with wal_consistency_checking and 13~  (Michael Paquier <michael@paquier.xyz>)
Responses Re: Failures with wal_consistency_checking and 13~
List pgsql-hackers
Em seg., 15 de jun. de 2020 às 10:14, Michael Paquier <michael@paquier.xyz> escreveu:
Hi all,

I have begun my annual study of WAL consistency across replays, and
wal_consistency_checking = 'all' is pointing out at some issues with
at least VACUUM and SPGist:
FATAL:  inconsistent page found, rel 1663/16385/22133, forknum 0,
blkno 15
CONTEXT:  WAL redo at 0/739CEDE8 for SPGist/VACUUM_REDIRECT: newest
XID 4619

It may be possible that there are other failures, I have just run
installcheck and this is the first failure I saw after replaying all
the generated WAL on a standby.  Please note that I have also checked
12, and installcheck passes.
With Postgres 13, Windows 10 (home), msvc 2019 64 bits,
Shutting down, without interrupting the database, consistently, this log has appeared.

2020-06-15 21:40:35.998 -03 [3380] LOG:  database system shutdown was interrupted; last known up at 2020-06-15 21:36:01 -03
2020-06-15 21:40:37.978 -03 [3380] LOG:  database system was not properly shut down; automatic recovery in progress
2020-06-15 21:40:37.992 -03 [3380] LOG:  redo starts at 0/8A809C78
2020-06-15 21:40:37.992 -03 [3380] LOG:  invalid record length at 0/8A809CB0: wanted 24, got 0
2020-06-15 21:40:37.992 -03 [3380] LOG:  redo done at 0/8A809C78
 
regards,
Ranier Vilela

pgsql-hackers by date:

Previous
From: Andres Freund
Date:
Subject: Re: global barrier & atomics in signal handlers (Re: Atomicoperations within spinlocks)
Next
From: Jeff Davis
Date:
Subject: Re: hashagg slowdown due to spill changes