Re: BUG #17973: Reinit of pgstats entry for dropped DB can break autovacuum daemon
Will Mortensen <will@extrahop.com>
From: Will Mortensen <will@extrahop.com>
To: Andres Freund <andres@anarazel.de>
Cc: pgsql-bugs@lists.postgresql.org, Jacob Speidel <jacob@extrahop.com>
Date: 2023-06-15T20:35:17Z
Lists: pgsql-bugs
Thanks for the quick patch Andres! Jacob tested and that appears to fix our scenario. On Wed, Jun 14, 2023 at 3:30 PM Andres Freund <andres@anarazel.de> wrote: > > Hi, > > On 2023-06-14 14:57:44 -0700, Andres Freund wrote: > > If I am not, then we we should defer setting MyDatabaseId until after "Recheck > > pg_database", e.g. by setting 'dboid' for the in_dbname path. > > > > I think that would likely end up with *less* code than before, because we'd > > need less redundant code. > > > > > > It's a bit odd that we do the second lookup with dbname, rather the oid we > > previously looked up. > > > > > > Afaics there's no reason to look up MyDatabaseTableSpace before the recheck. > > Attached is a rough prototype implementing this idea. Could you check if that > fixes the issue? > > Greetings, > > Andres Freund
Commits
-
Fix handling of shared statistics with dropped databases
- 5a9325fdde1d 15.5 landed
- c397c98ae63d 16.0 landed
- 2b8e5273e949 17.0 landed
-
Accept fractional seconds in jsonpath's datetime() method.
- bd590d1fea1b 15.4 cited