pgsql: Avoid resetting Xmax when it's a multi with an aborted update
От | Alvaro Herrera |
---|---|
Тема | pgsql: Avoid resetting Xmax when it's a multi with an aborted update |
Дата | |
Msg-id | E1VobQw-0006wL-3z@gemulon.postgresql.org обсуждение исходный текст |
Ответы |
Re: pgsql: Avoid resetting Xmax when it's a multi with an aborted update
|
Список | pgsql-committers |
Avoid resetting Xmax when it's a multi with an aborted update HeapTupleSatisfiesUpdate can very easily "forget" tuple locks while checking the contents of a multixact and finding it contains an aborted update, by setting the HEAP_XMAX_INVALID bit. This would lead to concurrent transactions not noticing any previous locks held by transactions that might still be running, and thus being able to acquire subsequent locks they wouldn't be normally able to acquire. This bug was introduced in commit 1ce150b7bb; backpatch this fix to 9.3, like that commit. This change reverts the change to the delete-abort-savept isolation test in 1ce150b7bb, because that behavior change was caused by this bug. Noticed by Andres Freund while investigating a different issue reported by Noah Misch. Branch ------ master Details ------- http://git.postgresql.org/pg/commitdiff/07aeb1fec571920839822c11851d38ef48952d6c Modified Files -------------- src/backend/utils/time/tqual.c | 21 ++++++++++++++++---- .../isolation/expected/delete-abort-savept.out | 13 +++++------- 2 files changed, 22 insertions(+), 12 deletions(-)
В списке pgsql-committers по дате отправления: