Dear Guenter
thank you for the quick answer.
We also had the idea to plug an multimeter between mass (black) and
the +12V wire (yellow).
The voltage was good at approx 12V.
Finally, I figured out that ACPI returned an incorrect value to the
operating system.
Therefore, I did the following:
Step 1: Get voltage values from the BIOS
VCORE => 1.392 V
+3.3V => 3.304 V
+5V => 5.130 V
+12V => 12.105 V
Step 2: Get the raw data from the sys-filesystem
cat /sys/class/hwmon/hwmon1/in1_input
3304
cat /sys/class/hwmon/hwmon1/in2_input
5130
cat /sys/class/hwmon/hwmon1/in3_input
10543
Step 3: Calculate values for sensors.conf
3304 / 3304 = 1.0
5130 / 5130 = 1.0
12105 / 10543 = 1.14816
So only 12V needs a correction.
Step 4: Created file /etc/sensors.d/atk0110-acpi.conf
chip "atk0110-acpi-*"
compute in3 @ * 1.14816, @ / 1.14816
I am still unsure why other machines with the same mainboard provide
correct values.
If I find a free time slot, I will check if all boards have the same
BIOS version and settings.
At the moment, the solution above is ok for me.
Thank you for your help.
Best regards
Sascha
Zitat von Guenter Roeck <linux@xxxxxxxxxxxx>:
On Wed, Sep 18, 2013 at 11:01:56AM +0200, Sascha Haupt wrote:
Dear @ll,
I am monitoring multiple systems (with Asus M5A78LE mainboards)
using sensors and Munin.
On one of the nodes, the +12 V value is constantly reported to be
+10.48 Volt which
causes Munin to report a warning (value outside 10.92 to 13.08 Volt range).
The value seems to be close to +12 V on all other nodes.
One idea was that the system reduces voltage to save power.
Therefore I caused some load on the node:
root:/home/root # md5sum /dev/urandom &
root:/home/root # md5sum /dev/urandom &
root:/home/root # md5sum /dev/urandom &
root:/home/root # md5sum /dev/urandom &
What I can see is that the 1.4 V value is now constanly at 1.38 V.
Previously, it jumped between 1.01 V and 1.38 V.
Makes sense, as the CPU voltage changes with load.
However, there is no difference in the +12 V value.
12V voltage does not usually change baed on power states.
I do not understand why the 12 V value stays at such an low voltage.
May any power saving options cause that behaviour (when yes, which
settings can cause that)?
Do you have any ideas how I can localise the cause of the +10.48 Volt value?
Maybe there is a problem with that system ? Can you measure the
voltage at the
power supply ? That may even be related to the fan problem you
reported earlier
if that is on the same system, as the fans operate on 12V.
Guenter
Thanks for your help.
Best regards
Sascha Haupt
_______________________________________________
lm-sensors mailing list
lm-sensors@xxxxxxxxxxxxxx
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
_______________________________________________
lm-sensors mailing list
lm-sensors@xxxxxxxxxxxxxx
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
