public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: R.E.Wolff@BitWizard.nl (Rogier Wolff)
To: khromy <khromy@lnuxlab.ath.cx>
Cc: "Stephen C. Tweedie" <sct@redhat.com>,
	linux-kernel@vger.kernel.org, ext3-users@redhat.com
Subject: Re: sync slowness. ext3 on VIA vt82c686b
Date: Wed, 3 Jul 2002 12:57:45 +0200 (MEST)	[thread overview]
Message-ID: <200207031057.MAA03204@cave.bitwizard.nl> (raw)
In-Reply-To: <20020703094031.GA4462@lnuxlab.ath.cx> from khromy at "Jul 3, 2002 05:40:31 am"

khromy wrote:
> On Wed, Jul 03, 2002 at 10:22:44AM +0100, Stephen C. Tweedie wrote:
> > Ugh.  My first guess would be that you have one enormously fragmented
> > filesystem.  13MB in 2 minutes?  A modern disk should get that amount
> > of data to disk in one second, but massive fragmentation can simply
> > kill disk performance.
> > 
> > If /home is on the same disk, do you get the same problem trying to
> > write there?
> 
> Yeah, /home/ is on the same disk.  Your guess might be right because
> that's what I was trying to show.  When I copy the file, which is in
> /home/(hda2) to /tmp/(hda1) and I sync, it takes almost 2 minutes.  But 
> if I copy the same file, which is in /home/(hda2) to /usr/local/(hda3),
> sync returns immediately.  This disk isn't that old either.

Get your data off that disk Immediately!

If you write a large file, ext2 will do a good job not fragmenting the
file. You should be able to get about 20M per second on a sequential
writes, about 10M per second, if your filesystem is badly fragmented.

So, the drive is taking abnormally long to read/write blocks. That is
an indication that it's "going to die soon".

That said, maybe there is a whole lot of (random) reads going on 
on that disk? Are you swapping at the same time? Or maybe your
dayly "updatedb" is running?

Roger. 

-- 
** R.E.Wolff@BitWizard.nl ** http://www.BitWizard.nl/ ** +31-15-2137555 **
*-- BitWizard writes Linux device drivers for any device you may have! --*
* There are old pilots, and there are bold pilots. 
* There are also old, bald pilots. 

  reply	other threads:[~2002-07-03 10:56 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-07-03  2:20 sync slowness. ext3 on VIA vt82c686b khromy
2002-07-03  2:57 ` khromy
2002-07-03  3:24 ` Andrew Morton
2002-07-03  3:35   ` khromy
2002-07-03  3:57     ` Andrew Morton
2002-07-03  4:03       ` khromy
2002-07-03  4:36         ` khromy
2002-07-03 16:29           ` Diego Calleja
2002-07-03 17:25           ` Diego Calleja
2002-07-03 17:43   ` Tomas Szepe
2002-07-03 18:19     ` khromy
2002-07-03 19:05       ` Tomas Szepe
2002-07-03 21:54       ` Rogier Wolff
2002-07-03  9:22 ` Stephen C. Tweedie
2002-07-03  9:40   ` khromy
2002-07-03 10:57     ` Rogier Wolff [this message]
2002-07-03 16:38       ` khromy
2002-07-03 11:43     ` Pedro M. Rodrigues
     [not found]     ` <Pine.LNX.4.33.0207031155080.7491-100000@coffee.psychology.mcmaster.ca>
2002-07-03 16:42       ` khromy
2002-07-03 10:25 ` Denis Vlasenko
2002-07-03  5:59   ` khromy

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=200207031057.MAA03204@cave.bitwizard.nl \
    --to=r.e.wolff@bitwizard.nl \
    --cc=ext3-users@redhat.com \
    --cc=khromy@lnuxlab.ath.cx \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sct@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox