You are not logged in.
Today I updated everything and later on restarted my desktop. Afterwards I'm unable to get sddm/xorg to work. I can get the computer to start in run level 3. When I start normally, without quiet mode it shows a bunch of things passing then the screen goes black except a little white dash int he top left and it seems to be frozen there. I assume that is where it's trying to start up xorg/sddm but freezes.
I checked journalctl for sddm errors and got:
-- Boot da555a5b4a4d43e6ae533a7703f68240 --
Apr 24 16:31:36 deathstar systemd[1]: Started Simple Desktop Display Manager.
░░ Subject: A start job for unit sddm.service has finished successfully
░░ Defined-By: systemd
░░ Support: https://lists.freedesktop.org/mailman/l … temd-devel
░░
░░ A start job for unit sddm.service has finished successfully.
░░
░░ The job identifier is 192.
Apr 24 16:31:36 deathstar sddm[607]: Initializing...
Apr 24 16:31:36 deathstar sddm[607]: Starting...
Apr 24 16:31:36 deathstar sddm[607]: Logind interface found
Apr 24 16:31:36 deathstar sddm[607]: Adding new display...
Apr 24 16:31:36 deathstar sddm[607]: Loading theme configuration from ""
Apr 24 16:31:36 deathstar sddm[607]: Display server starting...
Apr 24 16:31:36 deathstar sddm[607]: Adding cookie to "/var/run/sddm/{d16d2ec1-42e8-47ed-b1ab-1183603059d6}"
Apr 24 16:31:36 deathstar sddm[607]: Running: /usr/bin/X -nolisten tcp -background none -seat seat0 vt2 -auth /var/run/sddm/{d16d>
Apr 24 16:31:37 deathstar sddm[607]: Failed to read display number from pipe
Apr 24 16:31:37 deathstar sddm[607]: Display server stopping...
Apr 24 16:31:42 deathstar sddm[607]: Attempt 1 starting the Display server on vt 2 failed
Apr 24 16:31:44 deathstar sddm[607]: Tried to start Xorg before previous instance exited
Apr 24 16:31:44 deathstar sddm[607]: Attempt 2 starting the Display server on vt 2 failed
Apr 24 16:31:46 deathstar sddm[607]: Tried to start Xorg before previous instance exited
Apr 24 16:31:46 deathstar sddm[607]: Attempt 3 starting the Display server on vt 2 failed
Apr 24 16:31:46 deathstar sddm[607]: Could not start Display server on vt 2
I also saw:
Apr 24 15:55:22 deathstar sddm[495]: Authentication error: "Process crashed"
Apr 24 15:55:22 deathstar sddm[495]: Auth: sddm-helper crashed (exit code 15)
Apr 24 15:55:22 deathstar sddm[495]: Authentication error: "Process crashed"
Apr 24 15:55:22 deathstar sddm[495]: Auth: sddm-helper exited with 15
Appears it cannot read the display number from pipe but I'm not not sure what that means exactly or if that is the right error.
the only error I really see in dmesg is
[ 3.285886] nvidia-gpu 0000:2d:00.3: i2c timeout error e0000000
[ 3.285893] ucsi_ccg 0-0008: i2c_transfer failed -110
[ 3.285896] ucsi_ccg 0-0008: ucsi_ccg_init failed - -110
[ 3.285903] ucsi_ccg: probe of 0-0008 failed with error -110
I saw people ask for kernel version and nvidia driver version on similar questions so:
pacman -Qs nvidia
local/cuda 11.3.0-1
NVIDIA's GPU programming toolkit
local/egl-wayland 1.1.6-1
EGLStream-based Wayland external platform
local/libnvidia-container 1.3.3-1
NVIDIA container runtime library
local/libnvidia-container-tools 1.3.3-1
NVIDIA container runtime library
local/libvdpau 1.4-1
Nvidia VDPAU library
local/nvidia 465.24.02-4
NVIDIA drivers for linux
local/nvidia-container-toolkit 1.4.2-1
NVIDIA container runtime toolkit
local/nvidia-utils 465.24.02-2
NVIDIA drivers utilities
local/opencl-nvidia 465.24.02-2
OpenCL implemention for NVIDIA
uname -a
Linux deathstar 5.11.16-arch1-1 #1 SMP PREEMPT Wed, 21 Apr 2021 17:22:13 +0000 x86_64 GNU/Linux
I tried running nvidia-smi and the computer froze up. So I'm not sure what else to debug or look for. I saw in another post someone said to setup early KMS but idk what that is: https://wiki.archlinux.org/index.php/Ke … _KMS_start. Like why set that up and how would it fix my issue? If anyone could help I would appreciate it. I also don't know how to format my text on this forum or if that is possible so sorry if the above looks u-g-l-y.
Edit 1:
Any time I probe the GPU everything seems to freeze. earlier when I tried nvidia-smi and now I just tried "torch.cuda.is_available()" in PyTorch to see if I could still use it for training models.
Last edited by Alom (2021-04-25 02:16:40)
Offline
Unfortunately, I'm on a time constraint so I ended up just reinstalling the OS. Was just easier for some reason.
Offline
Has anyone else been having this issue? Because I get that as well. Force downgraded the system to an earlier working state, but would pretty much be interested if somebody had figured out what's going on there.
Offline
Same probleme here !
I was able to reboot my computer after loading the kernel with the `rescue` parameter (see https://wiki.archlinux.org/index.php/ke … ters#GRUB) and downgrading the following packages with `pacman -U` (see: https://wiki.archlinux.org/index.php/do … _packages)
- nvidia-460.67-7-x86_64
- nvidia-settings-460-67-1-x86_64
- nvidia-utils-460.67-1-x86_64
- linux-5.11.13.arch1-1-x86_64
Not sure, but nvidia 460.67 does not seems to be compatible with linux 5.11.16.
Last edited by 4lgar (2021-04-27 10:14:11)
Offline