Re: corrupt pages detected by enabling checksums

Поиск
Список
Период
Сортировка
От Simon Riggs
Тема Re: corrupt pages detected by enabling checksums
Дата
Msg-id CA+U5nMK0K4zaeGaPLJdyJSWz+=h7azrawm9XzJApsxU1P=Q+xQ@mail.gmail.com
обсуждение исходный текст
Ответ на Re: corrupt pages detected by enabling checksums  (Robert Haas <robertmhaas@gmail.com>)
Список pgsql-hackers
On 11 April 2013 00:37, Robert Haas <robertmhaas@gmail.com> wrote:
On Sat, Apr 6, 2013 at 10:44 AM, Andres Freund <andres@2ndquadrant.com> wrote:
> I feel pretty strongly that we shouldn't add any such complications to
> XLogInsert() itself, its complicated enough already and it should be
> made simpler, not more complicated.

+1, emphatically.  XLogInsert is a really nasty scalability
bottleneck.  We need to move as much logic out of that function as
possible, and particularly out from under WALInsertLock.

Andres' patch was applied, so not sure what you mean by +1ing a comment made in relation to that patch.

I'm aware that WALInsertLock is a bottleneck and am not going to be making that worse.
 
--
 Simon Riggs                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Gurjeet Singh
Дата:
Сообщение: Re: [DOCS] synchronize_seqscans' description is a bit misleading
Следующее
От: Ants Aasma
Дата:
Сообщение: Re: Inconsistent DB data in Streaming Replication