Re: [SLE] Cautionary Tales: lm_sensors for beginners


Basil Chupin wrote:
Carlos E. R. wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


The Saturday 2006-09-09 at 23:17 +1000, Basil Chupin wrote:

- because the lm_sensors file is not visible.
Make it 'visible' to SUSE and the sound starts.

That may be because the script/program responsible does not run, ie,
exits, if it doesn't find the configuration file.

I feel that we are starting to lose the plot here.

I posted my original message for the simple reason that by having the
lm_sensors file created a high-pitched noise started up when I booted
into SUSE. It took a while to locate the reason for this noise so I
posted the message in case someone else, at some future time, strikes
the same problem and wonders what may be the sudden cause of the noise.

Joe Morris responded with the statement that nothing happens with
lm_sensors until one copies the startup script from
/usr/share/doc/packages/sensors/prog/init to /etc/init.d and then
enables the script; he asked if I did this and I answered NO.

By accepting that this script had to be activated before lm_sensors
started to actually work, I suggested that my (original) message was
more relevant than I initially thought because here was a control file
causing noise when the script which used it wasn't even activated (and
that when this file was removed the noise stopped).

While I still don't really know how lm_sensors work or what makes it
work, after reading quickly the wording of the lm_sensors file created
by /usr/sbin/sensors-detect command I now think that in SUSE 10.1 the
sensors start working when the file lm_sensors exists in
/etc/sysconfig - and this goes contrary to what Joe suggested.

If this is the case, that the sensors app. runs when lm_sensors is
present in /etc/sysconfig, then it makes perfect sense why the noise
stops when I rename the lm_sensors file - and what you state above is
correct.

But we got to this point after tossing some ideas around :-) .
However, I hope that my original posting, about what may be causing
the sudden appearance of a high-pitched noise when fooling around with
lm_sensors, comes in handy to someone in the future. The only thing we
haven't resolved - but does it matter? - is what exactly in the
parameters created by sensors-detect is causing the noise to be
generated? :-)

Cheers.

I very happy to see Basil refocus this discussion on his original issue.
For one, I am quite interested in learning something of value from this
discussion and
I am having trouble wading through all the words for the meat. Sorry!

My understand of lm_sensors (which could easily be very wrong) is that
it reads out from
various registers on the MBO values and then displays them for us. I
was always under the
impression that it is read only. How can a read only program cause a
change that sets
off Basil's alarm? Someone smart please explain that?

In addition, Steve Boddy in a post to this thread said:

This may sound like an silly question, but is it possible it's an alarm?

I've always found the lm_sensors stuff a bit hit and miss for detecting and
setting things properly, like putting odd values in for alarm conditions.

Not understanding the detail of how it works, it may be seeing something it
doesn't like and generating an audible alarm.

w83697hf-isa-0290
Adapter: ISA adapter
VCore: +1.66 V (min = +1.71 V, max = +1.89 V) ALARM
+3.3V: +1.49 V (min = +3.14 V, max = +3.47 V) ALARM
+5V: +4.87 V (min = +4.76 V, max = +5.24 V)
+12V: +12.04 V (min = +10.82 V, max = +13.19 V)
-12V: +1.54 V (min = -13.18 V, max = -10.80 V) ALARM
-5V: +0.13 V (min = -5.25 V, max = -4.75 V) ALARM
V5SB: +5.46 V (min = +4.76 V, max = +5.24 V) ALARM
VBat: +3.15 V (min = +2.40 V, max = +3.60 V)
fan1: 2766 RPM (min = 84375 RPM, div = 8) ALARM
fan2: 0 RPM (min = 1739 RPM, div = 8) ALARM
temp1: +38 C (high = +34 C, hyst = -92 C) sensor = thermistor
ALARM
temp2: +46.5 C (high = +100 C, hyst = +95 C) sensor = thermistor
alarms:
beep_enable:
Sound alarm enabled

As you can see, after a quick detect and init, I've got crazy values in some
fields, and alarms up the kazoo. Note also the beep_enable at the end. I
guess if that was wired up properly I'd be being annoyed right now by a high
pitched noise ;-)

I also have these alarms from the voltages being out of range from the norm.
However, it is clear to me, (maybe I am stupid) that these voltages can't be
incorrect as the BIOS looks fine and the machine works perfectly. Have
any of
you found what the right method to edit these files for accuracy? I
suppose i could
get the MBO manuals out and try to find the proper numbers from the
manufacture.
Seems a bit arcane though.

Again, my goal would be to learn and document more about the use of
lm_sensors
from as many smart people as I can for current and future reference.
Hopefully no
one reacts negatively to this, I just want to get smarter if possible.
And yes, I have
done a lot reading and note taking from the man pages and other DOC.

Cheers,
Bob


--
Check the headers for your unsubscription address
For additional commands send e-mail to suse-linux-e-help@xxxxxxxx
Also check the archives at http://lists.suse.com
Please read the FAQs: suse-linux-e-faq@xxxxxxxx

Relevant Pages

  • Re: [SLE] Cautionary Tales: lm_sensors for beginners
    ... really high tech and start using lm_sensors on SUSE 10.1. ... Took a while to pinpoint the sound as coming from the rheostat on the ... control knob gets rid of the noise. ... like putting odd values in for alarm conditions. ...
    (SuSE)
  • Re: [SLE] Cautionary Tales: lm_sensors for beginners
    ... really high tech and start using lm_sensors on SUSE 10.1. ... Took a while to pinpoint the sound as coming from the rheostat on the ... control knob gets rid of the noise. ... I've always found the lm_sensors stuff a bit hit and miss for detecting and setting things properly, like putting odd values in for alarm conditions. ...
    (SuSE)
  • Re: Palm Desktop & Entourage->Address Book/iCal/Mail.app
    ... of everything in my Palm Desktop), there are many, many To-Do items. ... alarm set for the due date, or no reminder alarm at all. ... We can change the script to just do one or the other (set alarm, ...
    (comp.sys.mac.apps)
  • RE: WebDav Worm?
    ... Perhaps urlscan could calm down the noise by keeping the infected host from sending the complete exploit by denying the SEARCH command. ... thought it was a script kiddie probing for various ... offsets/length of NOP sleds, ...
    (Incidents)
  • Re: Palm Desktop & Entourage->Address Book/iCal/Mail.app
    ... So set it to "don't set alarm." ... I added a property at the very top of the script that lets you choose ... I thought I could quit the script application from ... drag the file onto its icon). ...
    (comp.sys.mac.apps)