[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: What's causing my load average?



On Fri, May 10, 2002 at 01:51:06PM +0100, Paul Sargent wrote:
> I've got a box here which is cliaming a load average of about 10, but as 'ps
> aux' shows no process taking any CPU time. Any idea's what's going on?

You don't necessarily have to have processes taking up CPU time to have
a high load average. Processes stuck in the 'D' state (uninterruptible
sleep) will add a load of 1 each.

> root     12441  0.0  0.0  1556  812 ?        D    May09   0:00 /usr/sbin/automou
> root     12446  0.0  0.0  1548  772 ?        D    May09   0:00 /usr/sbin/automou
> root     12453  0.0  0.0  2072  996 ?        D    May09   0:00 in.rshd
> pauls    12525  0.0  0.0  2604 1316 ?        D    May09   0:00 -csh
> pauls    12535  0.0  0.0  2604 1312 ?        D    May09   0:00 -csh
> root     12550  0.0  0.0  2356 1412 ?        D    May09   0:00 -bash
> root     12603  0.0  0.0  2336 1304 ?        D    May09   0:00 -bash
> root     12893  0.0  0.0  1900  944 ?        S    06:25   0:00 /USR/SBIN/CRON
> root     12894  0.0  0.0  2028  908 ?        S    06:25   0:00 /bin/sh -c test -
> root     12895  0.0  0.0  1252  348 ?        S    06:25   0:00 run-parts --repor
> root     12922  0.0  0.0  2028  924 ?        S    06:25   0:00 /bin/sh /etc/cron
> root     12923  0.0  0.0  2048  952 ?        S    06:25   0:00 /bin/sh /usr/bin/
> root     12937  0.0  0.0  2048  952 ?        S    06:25   0:00 /bin/sh /usr/bin/
> root     12938  0.0  0.0  2336  516 ?        S    06:25   0:00 sort -f
> root     12939  0.0  0.0  1244  316 ?        S    06:25   0:00 /usr/lib/locate/f
> nobody   12940  0.0  0.0  1388  556 ?        D    06:25   0:00 /usr/bin/find / (

It looks like your automounter has got stuck, perhaps not helped by
/etc/cron.daily/find (probably) trawling paths it shouldn't? Unsticking
the automounter may require a reboot if you're unlucky - after that,
make sure /etc/updatedb.conf ignores the right filesystems and/or paths.

> root     13090  0.0  0.0     0    0 ?        SW   12:52   0:00 [nfsd]
> root     13091  0.0  0.0     0    0 ?        SW   12:52   0:00 [nfsd]
[and 14 more]

That might not be great either. Perhaps your upstream NFS server has
gone away?

I try to mount NFS filesystems using the 'intr' mount option so that I
can interrupt processes that are trying to access them if something goes
wrong.

Regards,

-- 
Colin Watson                                  [cjwatson@flatline.org.uk]


-- 
To UNSUBSCRIBE, email to debian-user-request@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org



Reply to: