From: khali@linux-fr.org (Jean Delvare)
To: lm-sensors@vger.kernel.org
Subject: Thanks and setup problem
Date: Thu, 19 May 2005 06:25:27 +0000 [thread overview]
Message-ID: <20041219185603.3ef62cf1.khali@linux-fr.org> (raw)
In-Reply-To: <41AE99A2.8050402@swbell.net>
Hi Chris,
> As you can see the first CPU jumps 14 Degrees in one second. And the
> other drops 7 degrees in the same second. This makes no sense. The
> jumps are too far on a second by second basis and they go up and down.
This isn't surprising. Thermal diode react very fast to temperature
changes, and CPU core temperature changes very, very fast. Most probably
one "sensors" command ran on one CPU, the second happened to be on the
other one. This explains the asymmetry, and shows that the adm1021 chips
do in fact provide very accurate, real-time temperature readings (these
chips are built into your Xeon CPUs).
> Now over a long period of time (minutes) you can see that as the
> temperatures go up and down they do seem to rise on the average. So
> they are not completely bogus. The temperatures rising, bring me to
> the second issue. Within 5 minutes (and this is reproducible) my
> motherboard alarm sounds. I think it happens when the CPU temp shows
> a reading of 61C (remember this reading is probably invalid) which is
> across the max of 60 that is displayed. This is interesting since
> the log does not show any indication an alarm has sounded. And I
Well this is all OK, the alarm is expected to fire up in these
conditions (providing the chips are wired to do this). I agree that the
output of sensors should reflect that. That said, maybe Intel's ADM1021
clones do simply not have that feature (alarm bits).
> have change the ADM chips configuration to shut off the beep. But I
> don't even think it is an option for this module. I can stress the
> CPU for a long time without this module loaded and never trigger the
> MB alarm. But less than 5 minutes with this in and it triggers all
> the time.
The version of lm_sensors you use (2.8.1) is a bit old. Back then, our
drivers would reset/reprogram chips arbitrarily. This might be the
difference. Try with a newer version and loading the module should not
make any difference anymore.
At any rate, your setup is almost perfect, there's not much more you can
hope. Temperatures, voltages and fan speeds are all OK as far as I can
see.
--
Jean Delvare
http://khali.linux-fr.org/
prev parent reply other threads:[~2005-05-19 6:25 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-05-19 6:25 Thanks and setup problem Chris Lopeman
2005-05-19 6:25 ` Mark M. Hoffman
2005-05-19 6:25 ` Chris Lopeman
2005-05-19 6:25 ` Jean Delvare [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20041219185603.3ef62cf1.khali@linux-fr.org \
--to=khali@linux-fr.org \
--cc=lm-sensors@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.