Fix theoretical torn page hazard.
authorJeff Davis <jdavis@postgresql.org>
Thu, 10 Nov 2022 22:46:30 +0000 (14:46 -0800)
committerJeff Davis <jdavis@postgresql.org>
Fri, 11 Nov 2022 20:38:29 +0000 (12:38 -0800)
commitd6a3dbe14f98d867b2fc3faeb99d2d3c2a48ca67
treefc44079af292bab900ccbae6736ee09db28754bf
parent3eb8eeccbee31597c5962de10dcb3930d780cb19
Fix theoretical torn page hazard.

The original report was concerned with a possible inconsistency
between the heap and the visibility map, which I was unable to
confirm. The concern has been retracted.

However, there did seem to be a torn page hazard when using
checksums. By not setting the heap page LSN during redo, the
protections of minRecoveryPoint were bypassed. Fixed, along with a
misleading comment.

It may have been impossible to hit this problem in practice, because
it would require a page tear between the checksum and the flags, so I
am marking this as a theoretical risk. But, as discussed, it did
violate expectations about the page LSN, so it may have other
consequences.

Backpatch to all supported versions.

Reported-by: Konstantin Knizhnik
Reviewed-by: Konstantin Knizhnik
Discussion: https://postgr.es/m/fed17dac-8cb8-4f5b-d462-1bb4908c029e@garret.ru
Backpatch-through: 11
src/backend/access/heap/heapam.c