Re: VACUUM FULL results in deadlock

Поиск
Список
Период
Сортировка
От Manuel Rigger
Тема Re: VACUUM FULL results in deadlock
Дата
Msg-id CA+u7OA61ia8BSjFmyoOyEjTdEKMLq5OeFOL9ayeCcp5vaq0k1w@mail.gmail.com
обсуждение исходный текст
Ответ на Re: VACUUM FULL results in deadlock  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-bugs
I assume that you are talking about the corruptions. I'm still trying
to make them reproducible. So far, I could not reproduce them by
trying to repeatedly replay the recorded statements.

Best,
Manuel

On Fri, Jul 5, 2019 at 7:08 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>
> Alvaro Herrera <alvherre@2ndquadrant.com> writes:
> > I agree that you should not be running 32 full vacuums at once, much
> > less in a loop, but if you do, they shouldn't result in weird corruption
> > such as the ones reported.
>
> I agree that we probably should look closer rather than just writing this
> off.  However, without a more concrete report there's no way to look
> closer.
>
>                         regards, tom lane



В списке pgsql-bugs по дате отправления:

Предыдущее
От: Manuel Rigger
Дата:
Сообщение: Re: VACUUM FULL results in deadlock
Следующее
От: Peter Geoghegan
Дата:
Сообщение: Re: BUG #15896: pg_upgrade from 10-or-earlier: TRAP: FailedAssertion(»!(metad->btm_version >= 3)«