Re: Defaulting wal_sync_method to fdatasync on Linux for 9.1?

From: Andres Freund <andres(at)anarazel(dot)de>
To: pgsql-performance(at)postgresql(dot)org
Cc: Greg Smith <greg(at)2ndquadrant(dot)com>, Marti Raudsepp <marti(at)juffo(dot)org>
Subject: Re: Defaulting wal_sync_method to fdatasync on Linux for 9.1?
Date: 2010-10-31 23:10:28
Message-ID: 201011010010.28724.andres@anarazel.de
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Sunday 31 October 2010 20:59:31 Greg Smith wrote:
> Writes only are sync'd out when you do a commit, or the database does a
> checkpoint.
Hm? WAL is written out to disk after an the space provided by wal_buffers(def
8) * XLOG_BLCKSZ (def 8192) is used. The default is 64kb which you reach
pretty quickly - especially after a checkpoint. With O_D?SYNC that will
synchronously get written out during a normal XLogInsert if hits a page
boundary.
*Additionally* its gets written out at a commit if sync commit is not on.

Not having a real O_DSYNC on linux until recently makes it even more dubious
to have it as a default...

Andres

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Marti Raudsepp 2010-11-01 01:29:41 Re: Defaulting wal_sync_method to fdatasync on Linux for 9.1?
Previous Message Greg Smith 2010-10-31 19:59:31 Re: Defaulting wal_sync_method to fdatasync on Linux for 9.1?