Fix MVCC bug with prepared xact with subxacts on standby (#8152)

We did not recover the subtransaction IDs of prepared transactions when
starting a hot standby from a shutdown checkpoint. As a result, such
subtransactions were considered as aborted, rather than in-progress.
That would lead to hint bits being set incorrectly, and the
subtransactions suddenly becoming visible to old snapshots when the
prepared transaction was committed.

To fix, update pg_subtrans with prepared transactions's subxids when
starting hot standby from a shutdown checkpoint. The snapshots taken
from that state need to be marked as "suboverflowed", so that we also
check the pg_subtrans.

Discussion:
https://www.postgresql.org/message-id/6b852e98-2d49-4ca1-9e95-db419a2696e0%40iki.fi

NEON: cherry-picked from the upstream thread ahead of time, to unblock
https://github.com/neondatabase/neon/pull/7288. I expect this to be
committed to upstream in the next few days, superseding this. NOTE: I
did not include the new regression test on v15 and v14 branches, because
the test would need some adapting, and we don't run the perl tests on
Neon anyway.
This commit is contained in:
Heikki Linnakangas
2024-06-25 16:29:32 +03:00
committed by GitHub
parent 219e78f885
commit d502313841
4 changed files with 6 additions and 6 deletions

View File

@@ -1,5 +1,5 @@
{
"v16": ["16.3", "9837db157837fcf43ef7348be0017d3a2238cd27"],
"v15": ["15.7", "e22098d86d6c40276b6bd75c29133a33fb283ab6"],
"v14": ["14.12", "4c51945a6167ca06c0169e7a4ca5a8e7ffa3faba"]
"v16": ["16.3", "3bf9219f6ef5e943393e9430872e26184e92d1c6"],
"v15": ["15.7", "2092a6dcee794bb0cb17471bd964690dd7c7355f"],
"v14": ["14.12", "aa88bd536b48b22328aac748be0dcfff760135d0"]
}