Hi all, sorry if I post this the wrong place.
I have a laptop running mint with qtile which sometimes freezes. To the point where nothing responds and I need to kill it. I’ve tried: sudo journalctl But I don’t get any information which helps me.
Can anyone help to debug it?
It might not be a bad idea having btop with like 100ms refresh. Watch it while using it and see what the thermals are as everyone is suggesting. Or if any other cpu thing is being weird. I’ve had weird freezes before due to lack of power. Or being on a screwed up power plan.
What laptop is it?
I would look at these things first.
- Try another DE, something like XFCE. See if the problem persists. Sometimes swapping compositors or display managers can help too.
- Run memtest. Failing memory can definitely cause lock-ups.
- Lastly I’d look at graphics drivers. If you’re running Nvidia, switch from nouveau to the proprietary driver or vice-versa and see if that helps.
Great idea. I will try some time with XFCE. One of the challenges is that I don’t know how to trigger the crash, and sometimes it can be days, other times multiple times within an hour.
Boot memtest
Leave it to do it’s thing overnight. That will at least check for badly failing RAM.
I’ve run this on machines that I thought were ok, only to find… they weren’t.
Can you check
/var/log/syslog?Something like:
tail -f /var/log/syslogmay help out.That file usually has the answers. Its just hard to find the exact lines.
There are an absurd amount of this:
2026-02-06T01:41:23.698044+01:00 tuxedo rtkit-daemon[46122]: Warning: Reached burst limit for user '1000', denying request. 2026-02-06T01:41:23.909775+01:00 tuxedo rtkit-daemon[46122]: Supervising 1 threads of 1 processes of 1 users. 2026-02-06T01:41:23.910579+01:00 tuxedo rtkit-daemon[46122]: Supervising 1 threads of 1 processes of 1 users. 2026-02-06T01:41:23.913096+01:00 tuxedo rtkit-daemon[46122]: Warning: Reached burst limit for user '1000', denying request. 2026-02-06T01:41:24.124958+01:00 tuxedo rtkit-daemon[46122]: Supervising 1 threads of 1 processes of 1 users. 2026-02-06T01:41:24.125834+01:00 tuxedo rtkit-daemon[46122]: Supervising 1 threads of 1 processes of 1 users. 2026-02-06T01:41:24.128451+01:00 tuxedo rtkit-daemon[46122]: Successfully made thread 58065 of process 58064 owned by '1000' RT at priority 20. 2026-02-06T01:41:24.128489+01:00 tuxedo rtkit-daemon[46122]: Supervising 2 threads of 2 processes of 1 users.That seems to be there all the time, and also before crashing/freezing
If your machine is a Tuxedo laptop, this thread might interest you. Seems as though this user was hitting thermal limits and their laptop would freeze/poweroff to keep from dying.
+1 to this line on inquiry.
When I’ve managed to get a modern Linux desktop to freeze, I’ve had a bad power supply, or heat issues, each time.
Maybe something to do with your audio?
Maybe https://bugs.launchpad.net/debian/+source/rtkit/+bug/1547589
not sure.
Can you ctl+alt+4 to another terminal? Could you ssh from another machine?
No, I can’t access another tty during a freeze, unfortunately.
No, I can’t access another tty during a freeze, unfortunately.
That’s may be a hint!
The only times I’ve had the desktop freeze on Linux and the alternate terminals fail to respond, I had a hardware issue.
In one case, I was on a Raspberry Pi, and my power supply was not delivering clean enough power for the board.
In another case, my fan wasn’t connected properly and the motherboard was overheating.
Since you specifically mention qtile you should undo your customizations and see if that fixes your problem?
If not, you should look at the journal after reboot:
journalctl -bBut you’ll need to filter it.
Try
journalctl -b | grep -v rtkit-daemon, which will remove the masses of entries you bemoaned in another comment (AFAICS all syslog entries should also be in the journal anyhow).Very important:
Please make note of when the problem happened, and if your journal entries even go that far in time.Also, add
-rto show the log in reverse. If you want to look at previous boots other than the last you can specify like sojournalctl -b -2 -r
It might also be overheating. If you can monitor the temperature it might give you a clue.
I checked that at some point, and I don’t think that was the case, although sometimes during a crash the fans starts going crazy.
What are the machine’s specs? I had a similar issue on a Ryzen 5 3500U laptop before, but more recent kernels (6.8+) don’t exhibit that behavior.
maybe the ram gets full?
i’d have htop and dmesg -w running visibly and see if there’s any issue showing on the screen as it freezes
