nokyan
thermal sensor log to file mode
Is there an existing issue for this?
- [x] I have searched the existing issues
Is your feature request related to a problem? Please describe.
For hardware problem identification, i would like to track in file thermal evolution up to crash, to be able to read the log after reboot and see if it's linked to high temperature on some component (CPU or GPU mainly, but it can be chipset, RAM, or other).
Describe the solution you'd like
To not fill the disk for nothing, my ideal would be an option to log the thermal data every 10 seconds (or every second) for the last hour (and removing older data) if i use this option, it write log where i asked it to do, if not, it's not writing to disk.
And the thermal data logs could be read as text in log file, and why not viewed as graph by opening it with resources (put it's for extra shinny, it's not what's needed for my usecase)
Describe alternatives you've considered
No response
Additional context
No response
Hi, Resources does print GPU and CPU temperatures in its trace logs. Maybe you can parse Resources' trace output to get your desired outcome. I'm afraid I don't plan on including other, more sophisticated means to log statistics to console right now. Resources is primarily a graphical app, printing stats to console would be out of scope, I think.
Maybe you can parse Resources' trace output
Does trace activate by default ? Where can i found it ? I've don't seen anything about that in the readme nor in the app.
Maybe you can parse Resources' trace output
Does trace activate by default ? Where can i found it ? I've don't seen anything about that in the readme nor in the app.
You can run flatpak run --env=RUST_LOG=resources=trace net.nokyan.Resources in your terminal. This will enable trace output.
flatpak run --env=RUST_LOG=resources=trace net.nokyan.Resources 2>&1 | grep temperature >thermal.log
Do most of it. I lake timestamp, but i have chronology.
and :
flatpak run --env=RUST_LOG=resources=trace net.nokyan.Resources 2>&1 | grep temperature | some graph drawer
can probably reach my other feat request #496
Do you plan to include more thermal information ? mainboard/chipset temperature ? Disk/ssd temperature ?
Do you plan to include more thermal information ? mainboard/chipset temperature ? Disk/ssd temperature ?
Disk temperature should be doable. I don't know about mainboard temperature, I don't really see a use case for it to be honest.
For disk temperature : track the lake of airflow around the disk for heavy load (mostly for SSD) For chipset : find bad cooling setup creating crash (forget thermal glue, or fan error, or bad placement... occasioning abnormal high temperature)