hei
the log at http://karlsbakk.net/uptime.log.gz is a log create with
while true
do
uptime >> log
done
this shows the system load is somehow looping?!?
the system behaves well and is in production, but I don't really
understand what the kernel is up to. same numbers are reported by top.
sar and top etc reports no or little cpu and I/O load (<3%). The system
is used as a general tools server doing some webserver, nagios and mrtg
stuff. System is running 2.6.9-mm1.
please cc: to me as I'm not on the list
roy
On Friday 21 January 2005 10:32 am, Roy Sigurd Karlsbakk wrote:
This is not re-creatable here. As expected the tight loop pegs my CPU to 100
and we see the load avg increasing.
Either this was introduced in mm1, gentoo fixes it, or its an isolated
incident with Mr R. Karlsbakk.
Can I be of service in any other capacity kernel gurus?
bot403@eric MVS $ uname -a
Linux eric 2.6.9-gentoo-r9 #6 Wed Jan 19 18:37:07 CST 2005 i686 AMD Athlon(tm)
XP 2600+ AuthenticAMD GNU/Linux
--------------Tried to recreate OP output below------------
<~:12:55:44>
root@eric >while true; do uptime >> log; done
<~:12:57:37>
root@eric >uniq log
12:55:45 up 1 day, 18:14, 7 users, load average: 0.69, 0.55, 0.38
12:55:46 up 1 day, 18:14, 7 users, load average: 0.69, 0.55, 0.38
12:55:47 up 1 day, 18:14, 7 users, load average: 0.69, 0.55, 0.38
12:55:48 up 1 day, 18:14, 7 users, load average: 0.69, 0.55, 0.38
12:55:48 up 1 day, 18:14, 7 users, load average: 0.87, 0.59, 0.39
12:55:49 up 1 day, 18:14, 7 users, load average: 0.87, 0.59, 0.39
12:55:50 up 1 day, 18:14, 7 users, load average: 0.87, 0.59, 0.39
12:55:51 up 1 day, 18:14, 7 users, load average: 0.87, 0.59, 0.39
12:55:52 up 1 day, 18:14, 7 users, load average: 0.87, 0.59, 0.39
12:55:53 up 1 day, 18:14, 7 users, load average: 0.87, 0.59, 0.39
12:55:53 up 1 day, 18:14, 7 users, load average: 0.88, 0.60, 0.39
12:55:54 up 1 day, 18:14, 7 users, load average: 0.88, 0.60, 0.39
12:55:55 up 1 day, 18:14, 7 users, load average: 0.88, 0.60, 0.39
12:55:56 up 1 day, 18:14, 7 users, load average: 0.88, 0.60, 0.39
12:55:57 up 1 day, 18:14, 7 users, load average: 0.88, 0.60, 0.39
12:55:58 up 1 day, 18:14, 7 users, load average: 0.88, 0.60, 0.39
--snip to shorten, you get the idea steady increase--
12:56:55 up 1 day, 18:15, 7 users, load average: 0.95, 0.67, 0.43
12:56:56 up 1 day, 18:15, 7 users, load average: 0.95, 0.67, 0.43
12:56:57 up 1 day, 18:15, 7 users, load average: 0.95, 0.67, 0.43
12:56:58 up 1 day, 18:15, 7 users, load average: 0.95, 0.67, 0.43
12:56:58 up 1 day, 18:15, 7 users, load average: 0.96, 0.67, 0.43
12:56:59 up 1 day, 18:15, 7 users, load average: 0.96, 0.67, 0.43
--snip more avg .96 each second--
12:57:12 up 1 day, 18:15, 7 users, load average: 0.96, 0.68, 0.44
12:57:13 up 1 day, 18:15, 7 users, load average: 0.96, 0.68, 0.44
12:57:13 up 1 day, 18:15, 7 users, load average: 0.97, 0.69, 0.44
12:57:14 up 1 day, 18:15, 7 users, load average: 0.97, 0.69, 0.44
12:57:15 up 1 day, 18:15, 7 users, load average: 0.97, 0.69, 0.44
12:57:16 up 1 day, 18:15, 7 users, load average: 0.97, 0.69, 0.44
12:57:16 up 1 day, 18:16, 7 users, load average: 0.97, 0.69, 0.44
12:57:17 up 1 day, 18:16, 7 users, load average: 0.97, 0.69, 0.44
12:57:18 up 1 day, 18:16, 7 users, load average: 0.97, 0.69, 0.44
12:57:19 up 1 day, 18:16, 7 users, load average: 0.97, 0.69, 0.44
12:57:20 up 1 day, 18:16, 7 users, load average: 0.97, 0.69, 0.44
<~:12:57:40>
root@eric >
> hei
>
> the log at http://karlsbakk.net/uptime.log.gz is a log create with
--snip--
--
----------------------------------------
--EB
> All is fine except that I can reliably "oops" it simply by trying to read
> from /proc/apm (e.g. cat /proc/apm).
> oops output and ksymoops-2.3.4 output is attached.
> Is there anything else I can contribute?
The latitude and longtitude of the bios writers current position, and
a ballistic missile.
????????????????--Alan Cox LKML-December 08,2000
----------------------------------------