Maximum room temperature (Tyan S246x)
LT V. H. Walke
walke at usna.edu
Mon Apr 22 15:10:21 PDT 2002
Temperature monitoring for the dual processor Tyan board is mostly
working in lm_sensors 2.6.3 (March 22, 2002). See the tickets
referenced on the lm_sensors web page: http://www.netroedge.com/~lm78/
Unfortunately, you still have to go into bios on every boot to
initialize the monitoring chips, but everything seems to work.
Fortunately it seems progress is being made - a solution was posted on
April 11th (again, see the web page).
Good luck,
Vann
On Mon, 2002-04-22 at 17:58, Steven Timm wrote:
> On Mon, 22 Apr 2002 alvin at Maggie.Linux-Consulting.com wrote:
>
> >
> > hi ya manel
> >
> > if your systems have "health monitoring"...
> > check your bios to see what it thinks is your
> > current system temp and cpu temp...
> >
>
> Am I missing something here? The BIOS sensors can only
> tell you what the temperature is when the machine is
> effectively idle. A good CPU load can be good for an
> increase of 7-10 degrees C.
>
> >
> > you're cpu reliability/performance goes down
> > by 1/2 for every 10 degree C
> > - ie if it would have lasted 5 more years...
> > ( starting temp is "normal/nominal temp" as provided
> > ( by intel or amd that they provide a cpu warranty for 5 years
> > - if temp went up another 10deg C... its now 2.5 yrs
> > - it temp went up 20 degree.... its now 1.25yrs...
> > - or some silly guidelines like that..
> >
> > to test if the ambient temperature is too high...
> > - add a regular fan and blow air on it...
> >
> > - if the cpu temp drops significantly...
> > than its too hot in the room
> >
> > max cpu temp...
> > http://users.erols.com/chare/elec.htm
> > http://www.heatsink-guide.com/maxtemp.htm
> >
> > -- add lm_sensors to your "distro" to read the cpu temp
>
> Does anyone have success yet with making
> lm_sensors work on the Tyan 246x series of dual AMD motherboards?
>
> > and if it gets too high... shutdown the server
> > or at least dont do heavy computations on it
> > - add more fans ... and better air flow...
> >
> > c ya
> > alvin
> > http://www.Linux-1U.net/CPU .. more specs ...
> >
> >
> > On Mon, 22 Apr 2002, Manel Soria wrote:
> >
> > > I'm wondering what is the maximum reasonable ambient
> > > temperature to have in a cluster room. In our room
> > > with 72 nodes we have about 29-30 oC (84-86 oF).
> > > Is this too high ? Can this be the cause of hardware
> > > failures ?
> > >
> >
> > _______________________________________________
> > Beowulf mailing list, Beowulf at beowulf.org
> > To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
> >
>
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
--
----------------------------------------------------------------------
Vann H. Walke Office: Chauvenet 341
Computer Science Dept. Ph: 410-293-6811
572 Holloway Road, Stop 9F Fax: 410-293-2686
United States Naval Academy email: walke at usna.edu
Annapolis, MD 21402-5002 http://www.cs.usna.edu/~walke
----------------------------------------------------------------------
More information about the Beowulf
mailing list