I'm a Gentoo user who maintains an Artix system for educational purposes. I usually boot up Artix about once a week, and apply the latest updates (pacman -Syu).
Last Saturday the startup of the graphic desktop failed. The boot proceeded normally until SDDM fired up. Then, instead of presenting a login screen, the whole system just froze up.
Today I worked on it a bit. I was able to disable the SDDM service temporarily by commenting out all the contents of /etc/init.d/sddm. So I managed to log in as root and ran pacman-Syu, hoping this would fix the problem. It didn't.
I dug through the log files. Here's some stuff from /var/log/sddm.log:
[14:59:16.463] (II) DAEMON: Initializing...
[14:59:16.593] (II) DAEMON: Starting...
[14:59:16.594] (II) DAEMON: Logind interface found
[14:59:16.597] (II) DAEMON: Adding new display...
[14:59:16.757] (II) DAEMON: Loaded empty theme configuration
[14:59:16.759] (II) DAEMON: Xauthority path: "/var/run/sddm/xauth_JQcdFO"
[14:59:16.759] (II) DAEMON: Using VT 7
[14:59:16.759] (II) DAEMON: Display server starting...
[14:59:16.759] (II) DAEMON: Writing cookie to "/var/run/sddm/xauth_JQcdFO"
[14:59:16.759] (II) DAEMON: Running: /usr/bin/X -nolisten tcp -background none -seat seat0 vt7 -auth /var/run/sddm/xauth_JQcdFO -noreset -displayfd 16
[14:59:23.482] (EE) DAEMON: Failed to read display number from pipe
[14:59:23.482] (II) DAEMON: Display server stopping...
[14:59:23.482] (WW) DAEMON: Attempt 1 starting the Display server on vt 7 failed
[14:59:25.433] (II) DAEMON: Display server starting...
[14:59:25.433] (II) DAEMON: Writing cookie to "/var/run/sddm/xauth_JQcdFO"
[14:59:25.433] (II) DAEMON: Running: /usr/bin/X -nolisten tcp -background none -seat seat0 vt7 -auth /var/run/sddm/xauth_JQcdFO -noreset -displayfd 16
[14:59:25.520] (EE) DAEMON: Failed to read display number from pipe
[14:59:25.520] (II) DAEMON: Display server stopping...
[14:59:25.520] (WW) DAEMON: Attempt 2 starting the Display server on vt 7 failed
[14:59:27.433] (II) DAEMON: Display server starting...
[14:59:27.433] (II) DAEMON: Writing cookie to "/var/run/sddm/xauth_JQcdFO"
[14:59:27.434] (II) DAEMON: Running: /usr/bin/X -nolisten tcp -background none -seat seat0 vt7 -auth /var/run/sddm/xauth_JQcdFO -noreset -displayfd 16
[14:59:27.531] (EE) DAEMON: Failed to read display number from pipe
[14:59:27.531] (II) DAEMON: Display server stopping...
[14:59:27.531] (WW) DAEMON: Attempt 3 starting the Display server on vt 7 failed
[14:59:27.531] (EE) DAEMON: Could not start Display server on vt 7
So then I looked at /var/log/xorg.0.log. This appears to be the real problem.
[ 26.065] (II) Module glamoregl: vendor="X.Org Foundation"
[ 26.065] compiled for 1.21.1.16, module version = 1.0.1
[ 26.065] ABI class: X.Org ANSI C Emulation, version 0.4
[ 26.084] (EE)
[ 26.084] (EE) Backtrace:
[ 26.084] (EE) unw_get_proc_name failed: no unwind info found [-10]
[ 26.084] (EE) 0: /usr/lib/Xorg (?+0x0) [0x5569eb682d4d]
[ 26.085] (EE) unw_get_proc_name failed: no unwind info found [-10]
[ 26.085] (EE) 1: /usr/lib/libc.so.6 (?+0x0) [0x7f394c4d9600]
[ 26.085] (EE)
[ 26.085] (EE) Segmentation fault at address 0x0
[ 26.085] (EE)
Fatal server error:
[ 26.085] (EE) Caught signal 11 (Segmentation fault). Server aborting
[ 26.085] (EE)
[ 26.085] (EE)
Please consult the The X.Org Foundation support
at http://wiki.x.org
for help.
[ 26.085] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
[ 26.085] (EE)
[ 26.093] (EE) Server terminated with error (1). Closing log file.
I guess I'll also ask for help at the wiki.x.org web site. Any suggestions will be welcome. Thanks!
First pls try to downgrade mesa to the prev version.
artist
Hey David, it looks like the segmentation fault is tied to the glamoregl module, which often points to Mesa or GPU driver issues. Downgrading Mesa, as the Artist suggested, is a good idea since recent updates might have introduced incompatibilities. Also, check if your video driver package (like xf86-video-amdgpu or similar) was updated and try reverting it if needed. Running startx manually can help confirm if the issue is with Xorg itself or with SDDM. These steps can help isolate whether it’s a rendering backend issue or something deeper in the display stack.
Thanks for the suggestions. I decided to take a brute-force approach; I just reinstalled Artix in the same root partition. I had trouble with the "community" ISO file, so I went with a minimal KDE/Openrc version, which worked (although it needed a kick in the pants at the "installing bootloader" step). So I'll have to reinstall a lot of packages, but that shouldn't take long. I kept good notes.