[Om2008.9] runaway proc events/0 && automatic reboot

Matthias Apitz guru at unixarea.de
Fri Jan 16 11:19:33 CET 2009


El día Thursday, January 15, 2009 a las 04:41:13PM +0100, Korbinian Rosenegger escribió:

> Hi
> 
> On Thu, 2009-01-15 at 16:21 +0100, Matthias Apitz wrote:
> > my small script monitoring the events/0 proc is logging as well when the
> > runaway occures, date and uptime:
> > 
> > # cat runaway.log 
> > Wed Jan 14 01:20:04 CET 2009
> >  01:20:04 up 18:26,  0 users,  load average: 0.42, 0.63, 0.49
> > Thu Jan 15 14:49:19 CET 2009
> >  14:49:19 up 18:26,  0 users,  load average: 0.53, 0.76, 0.64
> > 
> > do you see the interesting point? the problem is uptime-driven!
> 
> 18 * 3600 + 26 * 60 = 66360
> 
> Interesting, since it is only ~14 Minutes above 65536 and your script
> checks about every 10 Minutes (i think). So maybe there's a unsigned int
> overflow or a similar problem somewhere?

after the yesterday re-boot at ~16pm the next runaway has just occured
just at the "scheduled" time at 10:17am:

Fri Jan 16 10:17:26 CET 2009
 10:17:26 up 18:17,  1 user,  load average: 0.27, 0.36, 0.33

this time the checking was done every 60 seconds which explains that it
was detected 9 minutes earlier after 18:17; it is safe to assume that
the runaway is not before 18:16, i.e.:

(18*3600)+(16*60)
65760

I will file this into the ticket
http://docs.openmoko.org/trac/ticket//1597

	matthias
-- 
Matthias Apitz
Manager Technical Support - OCLC GmbH
Gruenwalder Weg 28g - 82041 Oberhaching - Germany
t +49-89-61308 351 - f +49-89-61308 399 - m +49-170-4527211
e <matthias.apitz at oclc.org> - w http://www.oclc.org/ http://www.UnixArea.de/
b http://gurucubano.blogspot.com/

SPAMer of the year: Subject: Alle Software ist Deutsche Sprachen
>From: -40 % die Neujahrsaktion <GabrielleKelley at grungecafe.com>




More information about the community mailing list