You are not logged in.
Hi,
switched from Mint/Cinnamon (after 5 years) to Arch/Xmonad a month ago, so a bit new to all this, but things have been moving steadily and I'm quite happy. However, I cannot resolve this issue with Xmonad. Basically I don't know where to start, logs give me nothing.
I have read probably 20 threads about Xmonad freezing, tried 'pkill -CONT xmonad-x86_64-l', tried tricks from Xmonad FAQ but neither works.
I'm using Xmonad 0.15. Not using Xmobar, using Polybar instead. At least once a day, Xmonad freezes. Last few times I went into more detail and everything was frozen. So, not a situation described in some other issues where one window stays active and working. In the last two occasions, no window accepted mouse or keyboard input, switching workspaces was not possible. Although, in earlier occurrences, I'm quite sure at least workspace switching with the keyboard worked.
Mod+Q doe snot work. Shift+Mod+Q works, Xmonad is restarted but I lose all my work. 'xmonad --recompile' gives no errors but also doesn't resolve anything.
Any clues where to look for issues? Checked all log files I can think of and nothing. Currently on AwesomeWM, Ill see if it freezes, because right now I can't even be sure it's Xmonad's problem.
UPDATE: not 100% sure yet but awesome also froze! Differently, everything BUT one window was frozen. However, that window was Kitty running Vim with tmux. Error is not the same, but maybe Awesome just handles the problem differently. Since kitty uses GPU, maybe something between Kitty/Nvidia/Xorg is problematic. Will continue to experiment.
EDITED: Solution: you can read down below, but it turned out it was classic instability of xf86-video-intel. Uninstalling it resolved the problem. Note that I was unaware it was still active, so doublecheck what drivers are being used by your X if you ran into similar problems.
Last edited by psiho (2020-08-21 20:44:38)
Offline
How are you starting X11?
I would still recommend posting your journal, dmesg and X11 log after replicating the issue because even if you didn't see anything wrong, someone else might notice something.
Last edited by Irets (2020-08-17 19:23:23)
Offline
It takes time to reporduce error. But here it goes with another incident, this time in Awesome.
~/.xsession-errors
2020-08-18 16:02:55 E: awesome: Failed to load 'system-software-install-symbolic': Failed to open file “system-software-install-symbolic”: No such file or directory
stack traceback:
/usr/share/awesome/lib/gears/surface.lua:99: in function </usr/share/awesome/lib/gears/surface.lua:91>
(...tail calls...)
/usr/share/awesome/lib/naughty/core.lua:784: in function 'naughty.notify'
/usr/share/awesome/lib/naughty/dbus.lua:193: in function </usr/share/awesome/lib/naughty/dbus.lua:117>
** Message: 17:02:43.386: tray.vala:129: check updates
** Message: 17:02:43.389: tray.vala:159: 35 updates found
2020-08-18 17:02:52 E: awesome: Failed to load 'system-software-install-symbolic': Failed to open file “system-software-install-symbolic”: No such file or directory
stack traceback:
/usr/share/awesome/lib/gears/surface.lua:99: in function </usr/share/awesome/lib/gears/surface.lua:91>
(...tail calls...)
/usr/share/awesome/lib/naughty/core.lua:784: in function 'naughty.notify'
/usr/share/awesome/lib/naughty/dbus.lua:193: in function </usr/share/awesome/lib/naughty/dbus.lua:117>
** Message: 18:02:43.308: tray.vala:129: check updates
** Message: 18:02:43.312: tray.vala:159: 35 updates found
2020-08-18 18:02:56 E: awesome: Failed to load 'system-software-install-symbolic': Failed to open file “system-software-install-symbolic”: No such file or directory
stack traceback:
/usr/share/awesome/lib/gears/surface.lua:99: in function </usr/share/awesome/lib/gears/surface.lua:91>
(...tail calls...)
/usr/share/awesome/lib/naughty/core.lua:784: in function 'naughty.notify'
/usr/share/awesome/lib/naughty/dbus.lua:193: in function </usr/share/awesome/lib/naughty/dbus.lua:117>
[29840:29840:0818/182022.869412:ERROR:vaapi_wrapper.cc(511)] vaInitialize failed: unknown libva error
[29840:29840:0818/182022.879542:ERROR:sandbox_linux.cc(374)] InitializeSandbox() called with multiple threads in process gpu-process.... those last two lines I don't see in my log. Others are quite repetitive
dmesg last lines after some time of inactivity:
[62311.759971] GpuWatchdog[721]: segfault at 0 ip 000055d0118f6d58 sp 00007f8355cc43a0 error 6 in chromium[55d00d5e8000+7867000]
[62311.759989] Code: be 51 29 ff 48 8b 7d b0 48 8d 45 c0 48 39 c7 74 05 e8 dc f7 b4 fe 41 8b 84 24 e8 00 00 00 89 45 b0 48 8d 7d b0 e8 68 11 eb fb <c7> 04 25 00 00 00 00 37 13 00 00 64 48 8b 04 25 28 00 00 00 48 3b
[62311.760068] audit: type=1701 audit(1597767622.102:78): auid=1000 uid=1000 gid=1000 ses=2 pid=667 comm="GpuWatchdog" exe="/usr/lib/chromium/chromium" sig=11 res=1
[62311.816914] audit: type=1334 audit(1597767622.155:79): prog-id=13 op=LOAD
[62311.816981] audit: type=1334 audit(1597767622.155:80): prog-id=14 op=LOAD
[62311.817849] audit: type=1130 audit(1597767622.155:81): pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=systemd-coredump@0-29837-0 comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[62312.970479] audit: type=1131 audit(1597767623.309:82): pid=1 uid=0 auid=4294967295 ses=4294967295 msg='unit=systemd-coredump@0-29837-0 comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success'
[62313.041941] audit: type=1334 audit(1597767623.382:83): prog-id=14 op=UNLOAD
[62313.041945] audit: type=1334 audit(1597767623.382:84): prog-id=13 op=UNLOAD
[62314.025794] GpuWatchdog[1737]: segfault at 0 ip 0000557f1574e8ac sp 00007f8e0f907450 error 6 in brave[557f106ba000+8300000]
[62314.025845] Code: 89 de e8 17 9d 0a ff 80 7d c7 00 79 09 48 8b 7d b0 e8 08 1f 64 fe 41 8b 84 24 e0 00 00 00 89 45 b0 48 8d 7d b0 e8 e4 e8 58 fb <c7> 04 25 00 00 00 00 37 13 00 00 48 83 c4 48 5b 41 5c 41 5d 41 5e
[62314.025974] audit: type=1701 audit(1597767624.365:85): auid=1000 uid=1000 gid=1000 ses=2 pid=1630 comm="GpuWatchdog" exe="/usr/lib/brave-bin/brave" sig=11 res=1dmesg and journal basically say the same thing, Chromium and Brave segfaulted. Weird thing is that they both work, only window that doesn't work is Kitty with tmux/Vim
Killing all Kitty processes resolved this (killing browsers haven't). I'm pretty sure it is Kitty and/or Kitty with GPU drivers. WIll switch back to Xmonad and non-gpu terminal to see if it changes anything,
Last edited by psiho (2020-08-18 16:51:28)
Offline
So... switching from Kitty to Termite changed nothing. Back in Xmonad. It gives less logs but again last dmesg log is:
[100383.029545] GpuWatchdog[32307]: segfault at 0 ip 0000559b9ca95d58 sp 00007f9125f1d3a0 error 6 in chromium[559b98787000+7867000]... and .xsession-errors stil gives this at the end:
ERROR:sandbox_linux.cc(374)] InitializeSandbox() called with multiple threads in process gpu-process.BTW, using lightdm to start X.
Xmonad freezes quite differently than Awesome. It is completely froze, no workspace switching, no mouse or keyboard react other than mouse pointer and only outside current window (Chrome), above statusbar.
So Kitty is out of equation. Last stuff on my list is Chromium/Nvidia drivers. I'll stay at termite and switch from Chromium to Firefox now.
Offline
Ah, I'm stupid. My xf86-video-intel is still active! It seems something got me interrupted when installing Nvidia and while I did install it, I haven't configured it and uninstalled intel drivers.
This might be a classic intel drivers issue. I'll check
Offline
Yup, confirmed. It was a classic problem with xf86-video-intel. Uninstalling it resolved the issue. No freezes for two days.
Offline
Can I suggest the subject be edited? This definitely isn't an xmonad-specific problem.
Offline