Close race condition between datfrozen and relfrozen updates.

Noah Misch <noah@leadboat.com>

Commit: f65ab862e3b8d96c6886641155d9447bc73b5126
Author: Noah Misch <noah@leadboat.com>
Date: 2024-04-29T17:24:56Z
Releases: 17.0
Close race condition between datfrozen and relfrozen updates.

vac_update_datfrozenxid() did multiple loads of relfrozenxid and
relminmxid from buffer memory, and it assumed each would get the same
value.  Not so if a concurrent vac_update_relstats() did an inplace
update.  Commit 2d2e40e3befd8b9e0d2757554537345b15fa6ea2 fixed the same
kind of bug in vac_truncate_clog().  Today's bug could cause the
rel-level field and XIDs in the rel's rows to precede the db-level
field.  A cluster having such values should VACUUM affected tables.
Back-patch to v12 (all supported versions).

Discussion: https://postgr.es/m/20240423003956.e7.nmisch@google.com

Files

PathChange+/−
src/backend/commands/vacuum.c modified +16 −12

Discussion