Re: Perfomance Tuning - Mailing list pgsql-performance

From Bruce Momjian
Subject Re: Perfomance Tuning
Date
Msg-id 200308112216.h7BMGiF06162@candle.pha.pa.us
Whole thread Raw
In response to Re: Perfomance Tuning  ("scott.marlowe" <scott.marlowe@ihs.com>)
Responses Re: Perfomance Tuning
Re: Perfomance Tuning
List pgsql-performance
scott.marlowe wrote:
> On Fri, 8 Aug 2003, Andrew Sullivan wrote:
>
> > On Fri, Aug 08, 2003 at 09:40:20AM -0700, Jonathan Gardner wrote:
> > >
> > > Redhat puts ext3 on by default. Consider switching to a non-journaling FS
> > > (ext2?) with the partition that holds your data and WAL.
> >
> > I would give you exactly the opposite advice: _never_ use a
> > non-journalling fs for your data and WAL.  I suppose if you can
> > afford to lose some transactions, you can do without journalling.
> > Otherwise, you're just borrowing trouble, near as I can tell.
>
> I'd argue that a reliable filesystem (ext2) is still better than a
> questionable journaling filesystem (ext3 on kernels <2.4.20).
>
> This isn't saying to not use jounraling, but I would definitely test it
> under load first to make sure it's not gonna lose data or get corrupted.

That _would_ work if ext2 was a reliable file system --- it is not.

This is the problem of Linux file systems --- they have unreliable, and
journalled, with nothing in between, except using a journalling file
system and having it only journal metadata.

--
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 359-1001
  +  If your life is a hard drive,     |  13 Roberts Road
  +  Christ can be your backup.        |  Newtown Square, Pennsylvania 19073

pgsql-performance by date:

Previous
From: Josh Berkus
Date:
Subject: Re: Odd problem with performance in duplicate database
Next
From: Ron Johnson
Date:
Subject: Re: Odd problem with performance in duplicate database