git.postgresql.org Git - postgresql.git/commitdiff

git projects / postgresql.git / commitdiff

Fix corner case in autovacuum-forcing logic for multixact wraparound.

author Robert Haas <rhaas@postgresql.org>

2015年6月19日 15:28:30 +0000 (11:28 -0400)

committer Robert Haas <rhaas@postgresql.org>

2015年6月19日 15:28:30 +0000 (11:28 -0400)

Since find_multixact_start() relies on SimpleLruDoesPhysicalPageExist(),
and that function looks only at the on-disk state, it's possible for it
to fail to find a page that exists in the in-memory SLRU that has not
been written yet. If that happens, SetOffsetVacuumLimit() will
erroneously decide to force emergency autovacuuming immediately.

We should probably fix find_multixact_start() to consider the data
cached in memory as well as on the on-disk state, but that's no excuse
for SetOffsetVacuumLimit() to be stupid about the case where it can
no longer read the value after having previously succeeded in doing so.

Report by Andres Freund.

src/backend/access/transam/multixact.c patch | blob | blame | history

diff --git a/src/backend/access/transam/multixact.c b/src/backend/access/transam/multixact.c

index 516a89fa2ec22e5478f24b73bb39646f18e78c89..4daa5ae9b02ecde10d86ef50aa552bccda6b217f 100644 (file)

--- a/src/backend/access/transam/multixact.c

+++ b/src/backend/access/transam/multixact.c

@@ -2661,6 +2661,18 @@ SetOffsetVacuumLimit(bool finish_setup)

(errmsg("oldest MultiXactId member offset unknown")));

}

+ /*

+ * If we failed to get the oldest offset this time, but we have a value

+ * from a previous pass through this function, assess the need for

+ * autovacuum based on that old value rather than automatically forcing