Re: Random shutdowns - some new details

[Pavel Machek <pavel-AlSwsSmVLrQ@public.gmane.org>]

Hi!

> >In-kernel ACPI will happily call /sbin/halt for you in case of system 
> >overheat.
> >Watch /proc/acpi/thermal/*/*
> 
> Thank you for this information - when digging through the sources I was 
> able to locate the responsible lines.

So try to locate those "/sbin/halt" calling lines, and replace with
something like "logger 'I refuse to die'", and see what happens. If it
does not trigger, you know it is something else.

> To get behind these shutdowns I added some lines dealing with thermal 
> events, mainly cat'ting /proc/acpi/thermal_zone/THRx/temperature, to the 
> script /usr/sbin/acpid_proxy.
> 
> This script is called by acpid with the text of the occured event as 
> parameter (The SuSE-way of handling acpi-events).
> 
> Three days ago another shutdown occured, showing two things:
> 1) I made some mistake when extending acpid_proxy. But the temperatures 
> were logged.
> 2) The temperature logged are far below the trip-points for a shutdown. 
> (see the attached files)
> 
> The trip-point for my thermal zone THRS is 92C, the actual temperature 
> when the event occured was 46C (logged)
> 
> What does this mean?

No idea. [BTW nice system: two thermal zones, I never seen it before].

> Please mind the attached files:
> acpid - a cut from the acpid-logfile
> THRC-trip_points - /proc/acpi/thermal_zone/THRC/trip_points
> THRS-trip_points - /proc/acpi/thermal_zone/THRS/trip_points
> 

-- 
When do you have a heart between your knees?
[Johanka's followup: and *two* hearts?]


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf