mjhjmtu

How should I troubleshoot spontaneous turn off of my notebook running Debian Stretch? What happens:

1. Boot to Windows (dual boot system with windows and Debian on separate drives)


2. Reboot


3. Boot into Debian


4. After some time passes (can be 3 to 10 minutes but I have never observed it after less than 3 minutes after the boot of Linux), the notebook just turns off (it does not shutdown, just turns off).


5. I boot again to Debian


6. Runs solid for many hours without further problems.

The problem is repeatable.
last gives

user pts/0 :0 Wed Jun 13 07:50 - crash (00:21)
reboot system boot 4.9.0-6-amd64 Wed Jun 13 07:50 still running

which is strange in itself since at 7:50 there was NO real crash, it just booted and the turn off happened later. The last log entries before the crash from /var/log/messages:

Jun 13 07:51:39 debian NetworkManager[692]: <info> [1528890699.1216] dhcp6 (wlan1): state changed timeout -> done
Jun 13 07:55:41 debian colord[707]: failed to get session [pid 3726]: No such device or address
Jun 13 07:55:41 debian liblogging-stdlog: [origin software="rsyslogd" swVersion="8.24.0" x-pid="672" x-info="http://www.rsyslog.com"] rsyslogd was HUPed

and from /var/log/syslog (including initial message on the new boot)

Jun 13 08:00:21 debian gpg-agent[1557]: can't connect to the SCdaemon: IPC connect call failed
Jun 13 08:00:23 debian gpg-agent[1557]: can't connect to the SCdaemon: IPC connect call failed
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@
^@^@Jun 13 08:12:21 debian kernel: [ 0.000000] microcode: microcode updated early to revision 0x1c, 
date = 2015-02-26

At the time of turn off, the temperature on all cores is between 40C-50C. I have only seen this happen recently (last 2 weeks) and only on the boot immediately following the boot to Windows.

The notebook uses dual graphics cards. Bumblebee is not configured though and no special attention was paid to graphics card configuration in Debian. Windows obviously uses nVidia Optimus to switch between cards as necessary.

In the last 2 weeks the difference in my usage of the computer was running a program under windows that uses heavily the discrete nVidia graphics card. And since this is the only change of how I use the notebook that I am aware of, it is a prime suspect. But I have no idea how it could affect spontaneous turn off (overheating would come to mind but why would it not turn off in Windows during the heavy usage, why under linux after a few minutes running without heavy use). Note that the notebook turns off as well in the case I shutdown from windows and then start it back to Debian.

More importantly, I have no idea how to troubleshoot this. There are no logs that I can find that would point towards the trouble.