Discussion:
How to reset Chassis intrusion detection ALARM in sensors?
(too old to reply)
Jean-David Beyer
2004-06-22 16:06:07 UTC
Permalink
When I run sensors(1), it prints out the various voltages, fan speeds, and
temperatures of my system that uses a SuperMicro X5DP8-G2 motherboard. It
always says the following:

alarms: Chassis intrusion detection ALARM

Now I have been known to open the box, and that releases the chassis open
microswitch that is connected to the correct pins on the motherboard, so I
would expect to see this alarm sometimes.

It is my understanding that alarms are reset when they are read, however,
and when the chassis is closed up, the microswitch is clearly pressed so
the alarm should stop. It does not.

Am I likely to have configured something wrong? How am I supposed to reset it?
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 241939.
/( )\ Shrewsbury, New Jersey http://counter.li.org
^^-^^ 12:00:00 up 1:31, 4 users, load average: 4.22, 4.24, 4.18
Chiefy
2004-06-22 17:51:24 UTC
Permalink
Post by Jean-David Beyer
It is my understanding that alarms are reset when they are read, however,
and when the chassis is closed up, the microswitch is clearly pressed so
the alarm should stop. It does not.
Am I likely to have configured something wrong? How am I supposed to reset it?
I have a feeling you need to clear it from the BIOS setup.
--
I had to hit him, he was beginning to make sense.
Jean-David Beyer
2004-06-22 18:56:02 UTC
Permalink
Post by Chiefy
Post by Jean-David Beyer
It is my understanding that alarms are reset when they are read, however,
and when the chassis is closed up, the microswitch is clearly pressed so
the alarm should stop. It does not.
Am I likely to have configured something wrong? How am I supposed to reset it?
I have a feeling you need to clear it from the BIOS setup.
Do you mean I must reboot the whole machine just to clear that? Ugh!
I do not like that at all.
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 241939.
/( )\ Shrewsbury, New Jersey http://counter.li.org
^^-^^ 14:55:00 up 4:26, 3 users, load average: 2.04, 2.09, 2.15
Tim Buchanan
2004-06-23 02:50:25 UTC
Permalink
Post by Jean-David Beyer
Post by Chiefy
Post by Jean-David Beyer
It is my understanding that alarms are reset when they are read, however,
and when the chassis is closed up, the microswitch is clearly pressed so
the alarm should stop. It does not.
Am I likely to have configured something wrong? How am I supposed to reset it?
I have a feeling you need to clear it from the BIOS setup.
Do you mean I must reboot the whole machine just to clear that? Ugh!
I do not like that at all.
Maybe the actual switch is not aligned correctly, so that when you put the
case back on the switch doesn't activate?
--
When once you have tasted flight, you will forever walk the earth
with your eyes turned skyward, for there you have been, and there you
will always long to return.
Jean-David Beyer
2004-06-23 11:58:42 UTC
Permalink
Post by Tim Buchanan
Post by Jean-David Beyer
Post by Chiefy
Post by Jean-David Beyer
It is my understanding that alarms are reset when they are read, however,
and when the chassis is closed up, the microswitch is clearly pressed so
the alarm should stop. It does not.
Am I likely to have configured something wrong? How am I supposed to reset it?
I have a feeling you need to clear it from the BIOS setup.
Do you mean I must reboot the whole machine just to clear that? Ugh!
I do not like that at all.
Maybe the actual switch is not aligned correctly, so that when you put the
case back on the switch doesn't activate?
I just checked that by pressing the microswitch all the way in and holding
it there for about an hour. sensors runs every 15 minutes (cron job) and
it has not reset the alarm condition.
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 241939.
/( )\ Shrewsbury, New Jersey http://counter.li.org
^^-^^ 07:55:00 up 21:26, 4 users, load average: 4.31, 4.22, 4.17
Jean-David Beyer
2004-06-23 12:51:09 UTC
Permalink
Post by Jean-David Beyer
Post by Chiefy
Post by Jean-David Beyer
It is my understanding that alarms are reset when they are read,
however, and when the chassis is closed up, the microswitch is
clearly pressed so the alarm should stop. It does not.
Am I likely to have configured something wrong? How am I supposed to reset it?
I have a feeling you need to clear it from the BIOS setup.
Do you mean I must reboot the whole machine just to clear that? Ugh!
I do not like that at all.
I am not sure about that, though. In the motherboard instruction book is
an entire chapter on the BIOS. The BIOS claims to report

CPU 1 and CPU2 and SYSTEM temperature.
CPU 1 and CPU2 and Chassis 1, 2 fan speeds
VCore
3.3 STBY voltage
3.3 Vcc
5V Vcc
12V Vcc
1.8V Vcc
-12V Vcc

and only that.

Furthermore, it reports chassis 3 and chassis 4 fan speeds.
Note: it does not report the intrusion detector switch.

As far as I know, there is no 3.3 volt standby voltage. There certainly is
none in the power supply cable, nor does the SSI EPS12V Power Supply
Design Guide Revision 2.1 specify one. There is a 5V standby voltage.

OTOH, sensors reports

Wed Jun 23 08:00:00 EDT 2004
w83627hf-isa-0290
VCore 1: +1.44 V (min = +1.34 V, max = +1.47 V)
VCore 2: +3.31 V (min = +3.13 V, max = +3.45 V)
+3.3V: +3.24 V (min = +3.20 V, max = +3.45 V)
+5V: +4.99 V (min = +4.84 V, max = +5.24 V)
+12V: +11.97 V (min = +11.48 V, max = +12.58 V)
-12V: -11.83 V (min = -13.11 V, max = -11.41 V)
V5SB: +5.43 V (min = +4.84 V, max = +5.24 V)
VBat: +3.23 V (min = +2.40 V, max = +3.60 V)
CPU0 fan: 5818 RPM (min = 3000 RPM, div = 2)
CPU1 fan: 4041 RPM (min = 3000 RPM, div = 2)
fan3: 0 RPM (min = 3000 RPM, div = 2)
System: +42C (limit = +60C) sensor = thermistor
CPU0: +53.5C (limit = +60C, hysteresis = +50C) sensor = thermistor
CPU1: +51.0C (limit = +60C, hysteresis = +50C) sensor = thermistor
vid: +1.400 V
alarms: Chassis intrusion detection ALARM
beep_enable: Sound alarm disabled

Here, I have had to guess which voltage is which, which fan is which, and
which thermo sensor is which. I am pretty sure I have guessed correctly.
The sensor controller chip is actually w83627hf which definately accepts
three fan speed inputs. The power supply fan has tachometer output going
to chassis fan 1 connector, and the fans going to the other three are also
tachometer fans.
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 241939.
/( )\ Shrewsbury, New Jersey http://counter.li.org
^^-^^ 07:55:00 up 21:26, 4 users, load average: 4.31, 4.22, 4.17
Alexander Linkenbach
2004-06-24 16:15:53 UTC
Permalink
The way I understand chassis intrusion alert is that it does not tell
you that mereley that you chassis is open - no point you could see that
yourself, counldn't you? But that it tells you that the chassis has been
opened recently by someone else possibly; then it makes no sense if the
alert is reset by simply putting it back together, does it? So you
definately have to reset it in the BIOS. Wether there is a way of doing
it from inside the OS I do not know though. But there is nothing wrong
with it if it does not reset by simply pressing the microswitch again.
alex
Nick Landsberg
2004-06-24 18:06:43 UTC
Permalink
Post by Alexander Linkenbach
The way I understand chassis intrusion alert is that it does not tell
you that mereley that you chassis is open - no point you could see that
yourself, counldn't you? But that it tells you that the chassis has been
opened recently by someone else possibly; then it makes no sense if the
alert is reset by simply putting it back together, does it? So you
definately have to reset it in the BIOS. Wether there is a way of doing
it from inside the OS I do not know though. But there is nothing wrong
with it if it does not reset by simply pressing the microswitch again.
alex
This question gets down to one of requirements and environment.

Take a situation where a system is located in a remote
location which is usually unstaffed (e.g. the computers
in the hut at the base of the cellular antenna towers.)

If there was an intruder who opened the cabinet,
I would definitely want it so that the alarm was *NOT*
cleared by simply closing the chassis.

On the other hand, there are situations where it's
a transient condition, and it's OK to clear the alarm
(but I do think that re-booting to clear is definitely
over-kill).

[Obligatory War Story]
Some years ago, after the re-unification of Germany,
the former East Germany was being "wired" for high-tech
phone service (fiber and coax). The "pedestals"
on the street corners (from which the rest of the
block of residences was to be wired) had a "door open alarm."
When we did the performance analysis, we assumed
that the frequency of this alarm would be "in the
noise" in comparison to other alarms.

As it happened, the roads/streets were in such bad
shape that a passing truck's vibrations would
trigger the "door open/door close" alarm sequence
several times in a very few seconds. The
alarm surveillance system became overloaded
because of this when they were hooked up to
several thousand pedestals.
[end Obligatory War Story]

NPL
--
"It is impossible to make anything foolproof
because fools are so ingenious"
- A. Bloch
Loading...