F44 Random desktop freezing with new dual monitor setup

Ask Fedora
22

Interesting. I also get this on my Nvidia GPU system, but in my case it’s a DisplayPort connection, not HDMI. I haven’t really dug in deeper since it’s just a cosmetic thing at login time.

23

i also got this when i used display port. Unplug the cable and re-attach it. Then shut down and reboot. See if that fixes it. Can’t hurt to try.

24

SAME EXACT ISSUE HERE - System spec’s at the bottom. Basically 3-screen and the Alienware middle monitor’s icon freeze. No left or right lick but I can drag the mouse across it, but the tool bar still works.

My temp solution is to restart plasma → `
``kquitapp6 plasmashell
kstart plasmashell```

BTW: Do you have Joplin installed?

I ask because some of my diagnosis pointed to Joplin, but I am not convinced

I am also running a few ‘watchers’ in 2 terminals to try to fix this. I never should have bought the hardware and allowed the updates.

Now - I do snapshots before any update, a monthly Clonezilla and I push a daily backup of certain folders to my NAS

Eddie

=== SYSTEM ===
System:
Host: fedora Kernel: 7.1.9-200.fc44.x86_64 arch: x86_64 bits: 64
Desktop: KDE Plasma v: 6.7.4 Distro: Fedora Linux 44 (KDE Plasma Desktop
Edition)

=== CPU ===
CPU:
Info: 16-core model: AMD Ryzen 9 7950X bits: 64 type: MT MCP cache:
L2: 16 MiB
Speed (MHz): avg: 3674 min/max: 425/5883 cores: 1: 3674 2: 3674 3: 3674
4: 3674 5: 3674 6: 3674 7: 3674 8: 3674 9: 3674 10: 3674 11: 3674 12: 3674
13: 3674 14: 3674 15: 3674 16: 3674 17: 3674 18: 3674 19: 3674 20: 3674
21: 3674 22: 3674 23: 3674 24: 3674 25: 3674 26: 3674 27: 3674 28: 3674
29: 3674 30: 3674 31: 3674 32: 3674

=== GRAPHICS ===
Graphics:
Device-1: NVIDIA GB203 [GeForce RTX 5080] driver: nvidia v: 610.57.04
Device-2: Logitech C920 PRO HD Webcam driver: snd-usb-audio,uvcvideo
type: USB
Display: wayland server: X.org v: 1.21.1.24 with: Xwayland v: 24.1.13
compositor: kwin_wayland driver: X: loaded: nvidia unloaded: modesetting
gpu: nv_platform,nvidia,nvidia-nvswitch resolution: 1: 2560x1440~120Hz
2: 1920x1080~60Hz 3: 1920x1080~60Hz
API: EGL v: 1.5 drivers: nvidia,swrast
platforms: gbm,wayland,x11,surfaceless,device
API: OpenGL v: 4.6.0 compat-v: 4.6 vendor: nvidia mesa v: 610.57.04
renderer: NVIDIA GeForce RTX 5080/PCIe/SSE2
API: Vulkan v: 1.4.341 drivers: nvidia,llvmpipe surfaces: N/A
Info: Tools: api: clinfo, eglinfo, glxinfo, vulkaninfo
de: kscreen-console,kscreen-doctor gpu: nvidia-settings,nvidia-smi
wl: wayland-info x11: xdriinfo, xdpyinfo, xprop, xrandr
eddie@fedora:~/source-to-signal$ kscreen-doctor -o | grep -E ‘Output:|enabled|connected|Geometry|Scale|Refresh|EDID|Name’
Output: 1 DP-2 c9cd31d4-2fdd-4688-8f23-1a9e9f011f35
enabled
connected
Geometry: 4480,200 1920x1080
Scale: 1
Output: 2 DP-1 89948cd1-32d9-446b-b1ae-ff21d0965379
enabled
connected
Geometry: 1920,0 2560x1440
Scale: 1
Output: 3 HDMI-A-1 ebd3ba3a-df22-445f-9a51-a7466ca41c8e
enabled
connected
Geometry: 0,200 1920x1080
Scale: 1

25

Before the freeze

  • Open terminal 1 and start the user-session watcher:
journalctl --user -f \
  | grep --line-buffered -Ei 'kwin|plasma|wayland|kscreen|qt|mouse|grab|drm|gpu|joplin' \
  >> ~/kwin-freeze-watch/user-live.log
  • Open terminal 2 and start the kernel/NVIDIA watcher:
sudo journalctl -kf \
  | grep --line-buffered -Ei 'nvidia|drm|gpu|xid|timeout|reset|fault|hang' \
  >> ~/kwin-freeze-watch/kernel-live.log
  • Leave both terminals running.
  • Use the machine normally, including Joplin.

If the center monitor freezes

  • Do not restart Plasma or KWin yet.

  • Record the exact time:

date --iso-8601=seconds | tee ~/kwin-freeze-watch/FREEZE_TIME.txt
  • Capture the monitor/output state:
kscreen-doctor -o \
  > ~/kwin-freeze-watch/kscreen-at-freeze.txt
  • Capture KWin’s internal state:
qdbus org.kde.KWin /KWin org.kde.KWin.supportInformation \
  > ~/kwin-freeze-watch/kwin-support-at-freeze.txt
  • Capture the recent user journal:
journalctl --user -b --since "5 minutes ago" \
  | grep -Ei 'kwin|plasma|wayland|drm|nvidia|gpu|joplin' \
  > ~/kwin-freeze-watch/user-snapshot-at-freeze.txt
  • Capture the kernel/NVIDIA state:
sudo dmesg \
  | grep -Ei 'nvidia|drm|gpu|xid|timeout|reset|fault|hang' \
  > ~/kwin-freeze-watch/kernel-snapshot-at-freeze.txt
  • If Joplin was open, closed, or crashed around the incident:
coredumpctl list | grep -i joplin \
  > ~/kwin-freeze-watch/joplin-coredumps.txt
  • Wait about 15–20 seconds so the two live watchers capture the aftermath.

Recovery

  • For normal recovery, use your usual Plasma restart first.

  • If Plasma restart does not fix the center screen, then use:

kwin_wayland --replace &

Be aware that replacing KWin can close terminals, browser windows, and other Wayland applications.

After recovery

  • Do not overwrite the incident folder.
  • Rename it with the freeze timestamp before starting a new watch session.

Example:

mv ~/kwin-freeze-watch ~/kwin-freeze-watch-20260824-192952
mkdir ~/kwin-freeze-watch
26

Are you still experiencing this? As of the latest kernel update the problem has fixed itself for me. Not sure why because I have been operating under the assumption that this was an nvidia driver bug. I’m guessing some type of timing issue because there wasn’t a new driver in the update just a new kernel. The issue wasn’t addressed in the update notes either.

I could make it freeze, pre update, by locking the screens, letting the monitors sleep, and then logging back in. The alienware monitor would be frozen every time. My work around was to create a keyboard shortcut with the following command. It’s akin to unplugging and re-plugging the display port cable on the affected monitor (DP-1):

kscreen-doctor output.DP-1.disable && sleep 2 && kscreen-doctor output.DP-1.enable

Log in > hit shortcut and desktop would reset. A minor inconvenience. I only mention it in case you are still having the issue. But the recent update fixed it for me.