Re: vac truncation scan problems
От | Michael Paquier |
---|---|
Тема | Re: vac truncation scan problems |
Дата | |
Msg-id | CAB7nPqTMBd6=5i3ZOg9t6A0km4VZ=wNt4_rwOzPVm683-aQ=Qw@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: vac truncation scan problems (Jeff Janes <jeff.janes@gmail.com>) |
Ответы |
Re: vac truncation scan problems
|
Список | pgsql-hackers |
On Wed, Apr 1, 2015 at 2:18 AM, Jeff Janes <jeff.janes@gmail.com> wrote:
On Tue, Mar 31, 2015 at 1:28 AM, Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp> wrote:Hi, this is a bug in the commit 0d831389749a3baaced7b984205b9894a82444b9 .
It allows vucuum freeze to be skipped and inversely lets regular
vacuum wait for lock. The attched patch fixes it.
In table_recheck_autovac, vacuum options are determined as following,
> tab->at_vacoptions = VACOPT_SKIPTOAST |
> (dovacuum ? VACOPT_VACUUM : 0) |
> (doanalyze ? VACOPT_ANALYZE : 0) |
!> (wraparound ? VACOPT_NOWAIT : 0);
The line prefixed by '!' looks inverted.Thanks, it is obvious once you see it!
Nice catch, Horiguchi-san.
--
Michael
В списке pgsql-hackers по дате отправления: