Show us your conky’s

[Member Peter Linu]

Helpful

Up

0

::

Thanks Afghan,
Please give me the WHOLE directions to the file in which I put this.

Live-USB = zzz-IceWM-antiX21-runit-64-base: static persistence
VirtualBox= zzz-IceWM-antiX21-runit-64-base

[Member afghan]

Helpful

Up

0

::

Peter Linu wrote:

Thanks Afghan,
Please give me the WHOLE directions to the file in which I put this.

Just put anywhere you like in your conky configuration file. You’ll find it in your home directory as ~/.conkyrc

You can find lots of sample lines in that file.

Beware of people who knows nothing; they are the ones that won't know that they know nothing.

[Member Peter Linu]

Helpful

Up

0

::

thank you afghan.

Live-USB = zzz-IceWM-antiX21-runit-64-base: static persistence
VirtualBox= zzz-IceWM-antiX21-runit-64-base

[Member Peter Linu]

Helpful

Up

0

::

afghan,
The “CPU Temp” is now certainly there, thank you, however it sits at 0 and doesn’t move. How do I get it working?

Live-USB = zzz-IceWM-antiX21-runit-64-base: static persistence
VirtualBox= zzz-IceWM-antiX21-runit-64-base

[Member afghan]

Helpful

Up

0

::

Peter Linu wrote:

The “CPU Temp” is now certainly there, thank you, however it sits at 0 and doesn’t move. How do I get it working?

Do you have acpid running? That conky command ${acpitemp} is reading what acpid reports. On my antiX21/runit system running on a Panasonic CF-SZ6 notebook it’s reported in /sys/class/termal/thermal_zone0/temp. Now as I’m typing here that file reads 43800 which is actually 43.8C. My conky rounds it and displays it as 44C.

If you have acpid and acpi utility installed then you can try the command:
acpi -t
to show the reading from temperature sensors; or
acpi -V
to show everything available on your setup.

btw you can use -f or -k switches to display temperatures in Farenheit or Kelvin scales. Check the acpi manpages.

Ymmv depending on your hardware, BIOS, and OS setup. Some notebooks like Toshibas are notorious for having (intentionally?) buggy BIOSes that just scews things up when using with anything else but MS Windows.

Good luck.

Beware of people who knows nothing; they are the ones that won't know that they know nothing.

[Member sybok]

Helpful

Up

0

::

ACPI/BIOS: During the boot, you may sometimes see messages with error/warning regarding ACPI.
That is my experience, moreover the location of these files may differ across different systems (antiX, Ubuntu etc.).
That is why I have proposed a solution where the appropriate data-files are explicitly found and parsed, see my 1st post in:
https://www.antixforum.com/forums/topic/cpu-temperature/.
Once these files are found*, this is – IMHO – a foolproof solution.
If it did not work for you, that you should have described what information was missing to test it, what you have tried and why it did not work (error messages) or what was hard to understand.

I admit I did not state this reasoning behind finding the files in my post in the aforementioned thread.

BTW, this is a 3rd thread you use to ask the same question (the 1st was https://www.antixforum.com/forums/topic/conky-temp/ , the 2nd the already mentioned above).

[Moderator BobC]

Helpful

Up

0

::

He could try what was suggested this latest time, and check the temperature against the one on the IceWM system monitor

Attachments:

1. 

   screenshot-2022Aug12_003715.jpg
   

[Member calciumsodium]

Helpful

Up

0

::

afghan wrote:

This is the cpu temp fragment from my conkyrc:

CPU temp:${alignr}${acpitemp}°C

On some of my systems, the acpitemp function in conky does not work even though acpid is running.

In that case, what would work is one of the following within the .conkyrc file:

CPU temp: ${hwmon 0 temp 1}°C

or

CPU temp: ${hwmon 1 temp 1}°C

or 

CPU temp: ${hwmon 2 temp 1}°C

Update: I even have one system where both the acpitemp and hwmon functions don’t work.

In that case, what worked was:

CPU temp: ${platform coretemp.0/hwmon/hwmon0 temp 2}°C

You will have to try them out.

• This reply was modified 8 months, 3 weeks ago by calciumsodium. Reason: updated info

[Moderator caprea]

Helpful

Up

0

::

Indeed none of the above works on my lenovo laptop.
What works is this, it’s one line
${color white}CPU Temp ${color orange}${alignr}${exec sensors | sed -n '/(crit/p' | sed -n '/temp1/p' | cut -d" " -f 9 | sed 's/+//'}

[Anonymous]

Helpful

Up

0

::

Hi guys,
there is no general command for conky to read out the temperatures. After all, you don’t all have “the same computer”.

Try it out.
In former times it was done, acpitemp was already mentioned here. Further there are

sensors
or
sensors -u

To save some cpu one found the (Linux hardware monitoring kernel API) hwmon more suitable. But for this you first have to know where the modules are located.

e.g.

$ for m in /sys/class/hwmon/* ; do echo -n "$m = " ; cat $m/name ; done

or

$ find /sys/class/hwmon/* -exec echo -n "{}: " \; -exec cat {}/name \;

tells you.
Each interface has a file “name”. So I know which temp module it is.

Then the user just has to put it in his conky.

[Anonymous]

Helpful

Up

0

::

calciumsodium wrote:

afghan wrote:

This is the cpu temp fragment from my conkyrc:

CPU temp:${alignr}${acpitemp}°C

On some of my systems, the acpitemp function in conky does not work even though acpid is running.

In that case, what would work is one of the following within the .conkyrc file:

CPU temp: ${hwmon 0 temp 1}°C


or

CPU temp: ${hwmon 1 temp 1}°C

or 

CPU temp: ${hwmon 2 temp 1}°C

Update: I even have one system where both the acpitemp and hwmon functions don’t work.

In that case, what worked was:

CPU temp: ${platform coretemp.0/hwmon/hwmon0 temp 2}°C

You will have to try them out.

@calciumsodium,

if it’s always the same changing modules, like in my case, you can compensate that with an if_statement.
Then you don’t need to care anymore, because it doesn’t need any correction.
Example GPU

${goto 10}${color9} GPU:${color8}${if_existing /sys/class/hwmon/hwmon1/name radeon} ${hwmon 1 temp 1}°${else} ${hwmon 2 temp 1}°${endif}

[Anonymous]

Helpful

Up

0

::

caprea wrote:

Indeed none of the above works on my lenovo laptop.
What works is this, it’s one line
${color white}CPU Temp ${color orange}${alignr}${exec sensors | sed -n '/(crit/p' | sed -n '/temp1/p' | cut -d" " -f 9 | sed 's/+//'}

Hm, but you already know that the command is executed exactly once in conky, which is kind of pointless if you display the temperature?

Try this with an interval command with time behind it. 😉

[Moderator caprea]

Helpful

Up

0

::

Hi male, thanks for feedback and interest but no, it definitely works. It shows 34°C when the laptop boots up, after some while 49°C and the temperature rises to 56°C when a video is running e.g.
i actually thought that i probably picked it up from one of your conkies.

[Member afghan]

Helpful

Up

0

::

caprea wrote:

Indeed none of the above works on my lenovo laptop.
What works is this, it’s one line
${color white}CPU Temp ${color orange}${alignr}${exec sensors | sed -n '/(crit/p' | sed -n '/temp1/p' | cut -d" " -f 9 | sed 's/+//'}

If this gets executed every second or so (depending how your conky is set up) then monitoring temperature in this way may increase the temperature of the CPU… just kidding.

Long gone are the days when anyone could look up the official (or sometimes unofficial) hardware reference manual for any computer and find whatever memory addresses or interrupt values that you needed to read out or modify whatever you needed; all while waiting for your program to load from the cassette tape deck.

Beware of people who knows nothing; they are the ones that won't know that they know nothing.

[Member ModdIt]

Helpful

Up

0

::

I have two I5 quad cores on the 2500K. On my main box the only thing I found to work is below.
Sadly I am unable to credit the writer of the line, failed memory cells.
${goto 6}Core0:${alignr}${color #FA180A}${execi 30 sensors | grep 'Core 0' | awk '{print $3}' | sed 's/+//' | sed 's/\.0//g'} $color

Quad core count starts at 0, line is for first coretemp.
Each core has a sensor according to docs and readout definitely differs according to core load.

On my second box different I5 stepping this does not work.

Case temp is supported by my setup so
${goto 6}${voffset 4}Case Temp:${alignr}${color #FA180A}${acpitemp}°C

For my GPU below works fine but depends on Nvidia driver and n-vidia settings aplication.

${goto 6}${voffset 4}GPU Temp:${alignr}${color #FA180A}${exec nvidia-settings -tq '[gpu:0]/GPUCoreTemp'}°C $color

I am still looking for a recipe to show nvme temp, idealy without running anything as root.
After installing the nvme-cli package I can read device statistics with sudo nvme smart-log /dev/nvme0n1

At present it looks like the passive NVME cooling may be too effective, according to SanDisk the device needs
about 60C working temperature, present is 41.
I have high case airflow to keep passive graphics card temp below throttling limit.

• This reply was modified 8 months, 3 weeks ago by ModdIt.
• This reply was modified 8 months, 3 weeks ago by ModdIt.

[Author]

Posts