Sudden shutdown Debian Trixie
- Thread starter Max2207
- Start date
Intel core i7 gen 6 and Nvidia 950M. Gemini says it was electric shorts OMG IM broke. My fault after running owaspzap without any awareness. With Gemini trying to run with Intel graphice. But once again sudden shutdown
Welcome to linux.org, Max! I'm not as knowledgeable as some others here, but I'm confident you'll get some good help here soon.
I'm running Debian Trixie (13.5; KDE Plasma) and have good success with the Intel integrated graphics.
I agree with @piorunz that we're seeing Nvidia GPU errors. Typically the fix is changing the video driver.
I'm running Debian Trixie (13.5; KDE Plasma) and have good success with the Intel integrated graphics.
I agree with @piorunz that we're seeing Nvidia GPU errors. Typically the fix is changing the video driver.
Well hello my seniors. Thanks a lot for aproaching very fast. Any guide document for this typical err. Im working to dissambly the laptop an changing thermal paste. Anyway ITS full of dirt. I Dunno its Linux very aware of hardware or just bug. Anyway if the community this active. My worried decreased 30%. Trying to be SOC and cooked my laptop over 90° Celcius
These consideration occur to me on the issues:
The error with the nouveau "FAULT" messages suggests that the nouveau driver has tried to access some hardware address in memory, but failed. This doesn't necessarily mean that nouveau caused a "shutdown" of the computer.
For nvidia cards using the nouveau, there's a lot of nvidia firmware in the package: firmware-misc-nonfree, and it's best installed. One can install it, but better still is to add the non-free and non-free-firmware repositories to the debian sources list if not already in place. See here: https://wiki.debian.org/SourcesList. Look particularly at the Examples section.
It would be helpful if you could clarify what happens. A "shutdown" usually means the computer actually turns off with no processes running. That is quite different to the graphics card crashing and just making the machine unresponsive in the GUI. In that latter case of a graphics driver fault or crash, the rest of the machine usually, or might, still function in the background, but might be inaccessible because the keyboard is no longer responsive. What's the situation here?
The temperature at 90 degrees celsius looks quite ominous, and a temperature that high just 5 minutes after booting with a light load of usage could be implicated in a shutdown. Common temperatures for running a laptop for light usage are around half that temperature value. For heavy usage such as compiling or intensive video programming the temperatures can rise to the 80s or more but there was no suggestion of intensive usage in posts so far. Some cpus will shutdown the computer at temperatures that threaten them which may actually be set or controlled in the BIOS.
Temperature problems can be caused by numerous factors including dust, failed fans, thermal paste that's broken down, faulty power supply, and even some misbehaving software like linux's "watchdog", though that last one is unlikely because the software is pretty robust.
Although the issue may not be with nouveau, one can test whether or not the issue is just the graphics card driver, by booting the machine to a text prompt which will run the operating system without the GUI, and checking whether the machine will run as intended.
To boot to a text prompt, one can follow the instructions here: https://linuxconfig.org/how-to-set-kernel-boot-parameters-on-linux in the section on "temporary boot parameters". It describes how to set boot options (parameters). To boot to text mode, add the number 3 to the end of the linux kernel boot line. Then hit ctrl+x and the machine should boot to a text prompt when the user can log in, use the network and any text based apps.
All in all, I think the temperature is the first issue to address. After clearing and cleaning up the hardware, one can install the program: lm-sensors, if not already installed, which will provide detailed temperature info on the cpu and graphics card temperatures when the command:
sensors is run in a terminal. To become informed one can read the man pages for sensors and sensors-detect which are fairly clear. sensors-detect should be run first in the terminal, as root, and can be run in auto mode to do what it needs to. It only ever needs to be run that once. After that sensors will output temperatures pretty comprehensively.
Last edited:
CaffeineAddict
Well-Known Member
It is noveau driver issue not Nvidia GPU issue.
it's typical how you misinterpret the error.
I did not misinterpret anything.
a) Nvidia = problem, and yet another thread in the forums.
b) No nvidia = no problem.
Simple to understand correlation even primary school children.
Similar threads
