Re: broken tables on hot standby after migration on PostgreSQL 16 (3x times last month)

On Fri, May 17, 2024 at 1:18 PM Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com> wrote:
> pá 17. 5. 2024 v 18:02 odesílatel Peter Geoghegan <pg(at)bowt(dot)ie> napsal:
>> You've shown an inconsistency between the primary and standby with
>> respect to the heap tuple infomask bits related to freezing. It looks
>> like a FREEZE WAL record from the primary was never replayed on the
>> standby.
>
>
> It think is possible so broken tuples was created before upgrade from Postgres 15 to Postgres 16 - not too far before, so this bug can be side effect of upgrade

I half suspected something like that myself. Maybe the problem
happened *during* the upgrade, even.

There have been historical bugs affecting pg_upgrade and relfrozenxid.
Commit 74cf7d46 is one good example from only a few years ago.

--
Peter Geoghegan