Re: Datatypes and performance

Поиск
Список
Период
Сортировка
От Richard Huxton
Тема Re: Datatypes and performance
Дата
Msg-id 200307071649.11338.dev@archonet.com
обсуждение исходный текст
Ответ на Re: Datatypes and performance  ("Mattias Kregert" <mattias@kregert.se>)
Ответы Re: Datatypes and performance  (Bruce Momjian <pgman@candle.pha.pa.us>)
Список pgsql-general
On Monday 07 Jul 2003 4:11 pm, Mattias Kregert wrote:
> > > If I turn fsync on and then pull the power cord while a
> > > number of clients are doing lots of inserts/updates and stuff,
> > > will the fsync then guarantee that no data will be lost or
> > > corrupted?
> >
> > You are surely kidding, aren't you ?
> >
> > Karsten
>
> No. No kidding.
> Just to clarify, what I mean is: With FSYNC enabled, after a power failure,
> after "pg_ctl start" and replay of xact log etc; Are COMMITTED transactions
> guaranteed to be intact, and are UNCOMMITTED transactions guaranteed not to
> appear in the tables?
>
> If the answer is "yes", then I understand the use of FSYNC.
>
> If the answer is "no", then i don't see the point in using FSYNC at all.

The answer is "yes" providing:
1. Your hardware didn't suffer a failure during the outage.
2. Your disks don't like to the operating-system.

The second point is important - some disks have write-cache enabled and report
"done" when data is written to the cache not the platter. Google for
discussion.

--
  Richard Huxton

В списке pgsql-general по дате отправления:

Предыдущее
От: "Maksim Likharev"
Дата:
Сообщение: Re: Datatypes and performance
Следующее
От: Richard Huxton
Дата:
Сообщение: SQL Functions and plan time