Monitoring CPU temperature: mbmon shows 201 degrees C

[Achilleas Mantzios]

Στις Tuesday 22 July 2008 10:16:02 ο/η Manolis Kiagias έγραψε:
> Achilleas Mantzios wrote:
> > Στις Tuesday 22 July 2008 00:25:46 ο/η Tore Lund έγραψε:
> >   
> >> Achilleas Mantzios wrote:
> >>     
> >>> ...
> >>> Having said that, the issue with the temperature must not be my thing :(
> >>> after kldload coretemp, i get
> >>> [achix at panix ~]% sysctl -a | grep tempera
> >>> hw.acpi.thermal.tz0.temperature: 40,0C
> >>> dev.cpu.0.temperature: -1
> >>> [achix at panix ~]% 
> >>> The first always is stuck to 40 and dev.cpu.0.temperature to -1.
> >>>       
> >> Achillea, have you told us what CPU you have?  Manolis presumes you have
> >> an Intel, but I do not see this information anywhere in your posts.  If
> >> you have a recent AMD, try the port k8temp.
> >>     
> >
> > Sorry, i have a 
> > CPU: Intel(R) Pentium(R) 4 CPU 2.66GHz (2672.74-MHz 686-class CPU)
> >   Origin = "GenuineIntel"  Id = 0xf29  Stepping = 9
> >   Features=0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE>
> >   Features2=0x4400<CNXT-ID,xTPR>
> > So coretemp is not for me.
> >   
> 
> Definitely.
> 
> > While experimenting, i noticed the 1st and 3rd temperatures from mbmon to be updated in a fashion that seems
> > natural. 
> > [achix at panix ~]% mbmon
> > Temp.= 41.0, 201.0, 42.0; Rot.= 3443,    0,    0
> > Vcore = 1.50, 1.81; Volt. = 3.30, 5.08, 11.43, -11.74, -1.69
> >   
> 
> What chipset is the mobo based on? mbmon runs fine on my 865G and a 3Ghz 
> P4 CPU. You are probably correct, the middle temp may represent a sensor 
> that is not recognized, but the other readings seem normal.
> 

i'll let you know next time i open the case. Is there any reading from dmesg or sysctl that can reveal
that info?

> > I started to trust mbmon, and i think the 1st temp must be motherboard, while the 3rd CPU,
> > and indeed the first value varies betaeen 41-42 degrees, and the third value from 39, (~ 100% idle)
> > to 45 (0% idle). So i assume the above must be right.
> > Yesterday i had mbmon -t > mbmon.out running all night and the highest CPU temp was at 46 deg C,
> > while highest MB temp was at 43 deg Celsius (if the previous assumptions about the interpretation of the 
> > output of mbmon are correct).
> > Both high temps happened while running periodic daily at 03:00 (which increased my trust in those).
> > All that, was after i blew the box/case inside and closed the case.
> > If i trust those numbers and their interpretation then i must not have a temperature problem (anymore).
> > Lets see how the machine behaves.
> > There is always the other usual suspect from the memory department :)
> >
> >   
> 
> 
> For memory, I would suggest memtest86. For stressing the machine, try 
> math/mprime in torture mode. Watch the temperatures and make sure you 
> leave it running for a couple of hours and you don't get any errors. 
> Usually, if you have a termperature problem it will bail out in half an 
> hour or less.
> 

Memtest86 is good enough, i have used it on other machines. Thx for the math/mprime hint.

> 



-- 
Achilleas Mantzios