Cu163, backup->install->restore; cu165, fresh install: sensors nct6775 broken

Getting Started with IPFire
1

After restore, neither “HW Fan” nor “HW Volt” Graph displayed data anymore.

EDIT:
This concerns Chip `Nuvoton NCT6791D Super IO Sensors’
→ module nct6775 : running

Please c.f. Bug 12821 – Regression: sensors: NCT6791D

2

Hi,

this is a common thing if the hardware you are restoring a backup to is not identical to the hardware used before. Please refer to ŧhis wiki page for further information on this.

Thanks, and best regards,
Peter Müller

3

It was identical - of course. No jota changed.
1.) Backup taken via WEB Gui → “iso”
2.) Restore form “iso”

Kind regards
Manfred

4

Core Update 165 Development Build: master/c55f5c8e
Running “sensors-detect” succeeds - including the NUVOTON Super-IO;
“Copy prog/init/lm_sensors.init to /etc/init.d/lm_sensors
for initialization at boot time.”
Unfortunately,
. . . ll /etc/init.d/lm_sensors
. . . ls: cannot access ‘/etc/init.d/lm_sensors’: No such file or directory

. . . find / -name lm_sensors
. . . ./etc/sysconfig/lm_sensors

cat /etc/sysconfig/lm_sensors
. . . # Generated by sensors-detect on Fri Mar 25 09:00:02 2022
. . . # This file is sourced by /etc/init.d/lm_sensors and defines the modules to <-----!
. . . # be loaded/unloaded.
. . . #
. . . # The format of this file is a shell script that simply defines variables:
. . . # HWMON_MODULES for hardware monitoring driver modules, and optionally
. . . # BUS_MODULES for any required bus driver module (for example for I2C or SPI).
. . .
. . . HWMON_MODULES=“coretemp nct6775”

Reboot: to no avail

/var/ipfire/sensors/settings :
. . . still contains the old, former values, obviously via ‘restore’
. . . not being updated
WRONG - CORRECT MYSELF: it is “fresh data”:
. . . via ssh: # ll /var/ipfire/sensors/settings
. . . -rw-r–r-- 1 nobody nobody 3.8K Mar 25 13:01 /var/ipfire/sensors/settings

5

This is one of the directories not being backed up related to Bug12817

1 Like
6

Yes, Adolf, correct.
But why does the procedure ‘sensors-detect’ seemingly succeed - but without consequences ?
At least, I could not find any, yet.

7

I am not sure what you mean by that. Maybe you could explain what sort of consequences you are concerned about with the operation of sensors-detect.

sensors-detect is run by the collectd initscript and it pulls all the lm_sensors data together which is then fed to rrd for making any required graphs.

You seemed to be looking for a lm_sensors initscript but one does not exist for IPFire because that job is done by collectd.

8

Thanks for that background information.
Then we are back to my OP #1, asking:
Why neither “HW temp”, “HW Fan” nor “HW Volt” are being displayed?
AND:
Why did it still work all those updates up to cu163 at all, until re-install and restore ?

9

I don’t understand why not. Nothing was changed with either collectd or lm_sensors.
The last change to lm_sensors was 14th April 2021 and came into to CU156.

In my CU164 on my running production system I have all the hardware data graphs I had before. This is SSD temperature and the temperature of the four cores in my cpu.

I can’t test anything on this in my vm testbed as vm’s don’t have any sensors to get data for.

Maybe some hiccup occurred with the rrd graph working but I am not familiar enough with that to give any suggestions.
You could search about rrd and graphs in the forum to see if there is anything from previous threads and/or wait for someone more knowledgeable on this to provide some suggestions to look at.

10

Please, c.f. update in OP: