lkml.org 
[lkml]   [2004]   [Jun]   [16]   [last100]   RSS Feed
Views: [more markup]  [less markup]  [headers]  [forward] 
 
Messages in this thread
/
DateWed, 16 Jun 2004 17:45:51 +1000
FromNick Piggin <>
SubjectRe: PROBLEM: Heavy iowait on 2.6 kernels
Guy Van Sanden wrote:
> My machine is heavily used for all kinds of file serving (mainly nfs),
> but also samba.
> Next to that, it is my home-server, so it runs apache2 (tuned to server
> only few clients), imap (cyrus), postfix, bugzilla (mysql) and distcc
> (used only a few times a week).
> It replaces a FreeBSD system (PII-333) running the same except distcc.
> Under 
> 
> The disk system is just a regular IDE disk (udma5) (60GB) and one
> external drive over USB2 (160 GB).  The external drive is rather slow
> (20-30 MB/sec), so I disabled it during the tests.
> 
> The weird thing is that I see this problem too when only running bonnie.
> A friend of mine tried that too under 2.6.6, his iowait went up to
> 0.15%, mine to 99%.
> 

The CPU can very easily max out the disks of course. If bonnie is
doing IO to files much larger than memory, it wouldn't be surprising
for io-wait to get close to 100%. Possibly your friend was doing all
IO out of cache?

If you definitely have a performance problem, set "Kernel Debugging"
on in the "Kernel Hacking" menu, then set "Magic SysRq key" on. When
your system hits this iowait problem, press Alt+SysRq+T a couple of
times over a few seconds.

Then post the output of `dmesg -s 1000000`.
We'll see what is waiting where.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 14:03    [from the cache]
©2003-2008