I have successfully used Tumbleweed with two monitors for quite a while, but lately the monitor that is connected via HDMI randomly crashes. The crashes seem to occur only when I click on a widget on the panel, to open Konsole or Dolphin, etc. The crashes do not occur each time, but often enough to be a nuisance. When it happens the HDMI screen goes black.
To recover from the crash, I have to turn the monitor off and back on, usually several times.
This random crash only occurs when using Tumbleweed. There is no such problem when using Debian or Arch on the same computer.
Any ideas?
Here are the system details:
Operating System: openSUSE Tumbleweed 20241208
KDE Plasma Version: 6.2.4
KDE Frameworks Version: 6.8.0
Qt Version: 6.8.1
Kernel Version: 6.11.8-1-default (64-bit)
Graphics Platform: X11
Processors: 16 × AMD Ryzen 7 5700G with Radeon Graphics
Memory: 13,5 GiB of RAM
Graphics Processor: AMD Radeon Graphics
Manufacturer: Micro-Star International Co., Ltd.
Product Name: MS-7B89
System Version: 1.0
Immediately following such black screen occurrence, run “sudo dmesg | grep ‘cut here’” in Konsole. If command returns positive response, run “sudo dmesg | susepaste” and provide resulting URLs here. If a null response, run “sudo journalctl -b | grep aile” and paste the result here, unless another null response, in which case run “cat /var/log/Xorg.0.log | susepaste” or “susepaste /var/log/Xorg.0.log” and provide resulting URLs here. Also paste here please input/output from “inxi -GSaz”.
Just now my HDMI monitor crashed. Luckily I had the terminal open on the healthy monitor. Here are the messages that I received:
theber@murphy:~> sudo dmesg | susepaste
Pasted as:
https://paste.opensuse.org/fb657dc712a2
Failed to get path for session '3
4
5': No session '3
4
5' known
Graphics server not found. Copying to clipboard is not possible.
theber@murphy:~> sudo journalctl -b | grep aile
Dez 11 16:36:35 murphy nscd[1205]: 1205 stat failed for file `/etc/services'; will try again later: No such file or directory
Dez 11 16:36:35 murphy nscd[1205]: 1205 stat failed for file `/etc/netgroup'; will try again later: No such file or directory
Dez 11 16:36:35 murphy irqbalance[1183]: thermal: failed to receive messages.
Dez 11 16:36:35 murphy irqbalance[1183]: Failed to initialize thermal events.
Dez 11 16:36:36 murphy mysql-systemd-helper[1448]: 2024-12-11 16:36:36 0 [ERROR] feedback plugin: failed to retrieve the MAC address
Dez 11 16:36:36 murphy syncthing[1355]: 2024/12/11 16:36:36 failed to sufficiently increase receive buffer size (was: 208 kiB, wanted: 7168 kiB, got: 416 kiB). See https://github.com/quic-go/quic-go/wiki/UDP-Buffer-Sizes for details.
Dez 11 16:36:36 murphy syncthing[1355]: [5TX7J] INFO: listenerSupervisor@dynamic+https://relays.syncthing.net/endpoint: service dynamic+https://relays.syncthing.net/endpoint failed: Get "https://relays.syncthing.net/endpoint": dial tcp: lookup relays.syncthing.net on [2a02:908:2:a::1]:53: dial udp [2a02:908:2:a::1]:53: connect: network is unreachable
Dez 11 16:36:36 murphy syncthing[1355]: [5TX7J] INFO: listenerSupervisor@dynamic+https://relays.syncthing.net/endpoint: service dynamic+https://relays.syncthing.net/endpoint failed: Get "https://relays.syncthing.net/endpoint": dial tcp: lookup relays.syncthing.net on [2a02:908:2:a::1]:53: dial udp [2a02:908:2:a::1]:53: connect: network is unreachable
Dez 11 16:36:36 murphy syncthing[1355]: [5TX7J] INFO: listenerSupervisor@dynamic+https://relays.syncthing.net/endpoint: service dynamic+https://relays.syncthing.net/endpoint failed: Get "https://relays.syncthing.net/endpoint": dial tcp: lookup relays.syncthing.net on [2a02:908:2:a::1]:53: dial udp [2a02:908:2:a::1]:53: connect: network is unreachable
Dez 11 16:36:56 murphy sddm-helper[1822]: Failed to write utmpx: No such file or directory
Dez 11 16:36:56 murphy kwalletd6[1843]: kf.wallet.kwalletd: Wallet failed to get opened by PAM, error code is -9
Dez 11 16:36:56 murphy kded6[2026]: org.kde.libkbolt: Failed to connect to Bolt manager DBus interface:
Dez 11 16:36:57 murphy org_kde_powerdevil[2228]: Failed to create wl_display (No such file or directory)
Dez 11 16:36:58 murphy org_kde_powerdevil[2228]: org.kde.powerdevil: [DDCutilDetector]: Failed to initialize callback
Dez 11 16:36:58 murphy org_kde_powerdevil[2228]: org.kde.powerdevil: org.kde.powerdevil.chargethresholdhelper.getthreshold failed "Charge thresholds are not supported by the kernel for this hardware"
Dez 11 16:36:58 murphy org_kde_powerdevil[2228]: org.kde.powerdevil: org.kde.powerdevil.backlighthelper.brightness failed
Dez 11 16:36:58 murphy org_kde_powerdevil[2228]: org.kde.powerdevil: org.kde.powerdevil.chargethresholdhelper.getthreshold failed "Charge thresholds are not supported by the kernel for this hardware"
Dez 11 16:36:59 murphy keepassxc[2550]: YubiKey: Failed to establish PCSC context.
Dez 11 16:37:07 murphy kglobalacceld[1947]: kf.globalaccel.kglobalacceld: KKeyServer::xcbKeyPressEventToQt failed
Dez 11 16:37:36 murphy kglobalacceld[1947]: kf.globalaccel.kglobalacceld: KKeyServer::xcbKeyPressEventToQt failed
Dez 11 16:42:23 murphy fwupd[5308]: ERROR:tcti:src/tss2-tcti/tctildr.c:477:tctildr_init_context_data() Failed to instantiate TCTI
Dez 11 16:42:23 murphy fwupd[5308]: 15:42:23.689 FuEngine failed to add device /sys/devices/LNXSYSTM:00/LNXSYBUS:00/MSFT0101:00/tpm/tpm0: failed to initialize TPM library
Dez 11 16:42:24 murphy fwupd[5308]: 15:42:24.254 FuEngine failed to add device /sys/devices/pci0000:00/0000:00:02.1/0000:16:00.1/ata6/host5/target5:0:0/5:0:0:0/block/sr0: failed to open /dev/sr0: No medium found
Dez 11 16:51:35 murphy env[6517]: 002c:err:winediag:getaddrinfo Failed to resolve your host name IP
Dez 11 16:55:40 murphy systemd[1773]: app-wine\x2dPrograms\x2dQuick\x20Cribbage\x203.5\x2dQuick\x20Cribbage\x203.5@05734664404646e48c7935622e4ed6a1.service: Failed with result 'exit-code'.
Dez 11 17:08:39 murphy kglobalacceld[1947]: kf.globalaccel.kglobalacceld: KKeyServer::xcbKeyPressEventToQt failed
Dez 11 17:09:46 murphy kglobalacceld[1947]: kf.globalaccel.kglobalacceld: KKeyServer::xcbKeyPressEventToQt failed
I can’t recall seeing susepaste produce such errors. I wonder if it has anything to do with the “disconnected” display. Was its Konsole window originally opened in the disconnected display?
Did you get a similar failure trying to paste Xorg.0.log? I recognize nothing as clear evidence as to cause of the display disconnect, but inxi did clearly identify its disconnected status.
When next it happens, try saving the then current Xorg.0.log, then within a fresh Xterm or Konsole, or a fresh Plasma session, try again uploading it. paste.opensuse.org is frustrating sometimes. You can paste there via web browser if necessary, or to some other pastebin.
If you are unable to elicit other responses from this thread, I suggest to start a thread in KDE’s own forums, where the messages provided may be recognized, or someone else might be familiar with similar disconnects. There were updates to Mesa again in 20241210, so perhaps the issue would become history with your next dup.
You might also try opening an IceWM session instead of Plasma, in an attempt to elicit similar misbehavior. If you can, addressing the amd driver developers on IRC: irc://irc.oftc.net/radeon may be your best next action, or even if you cannot.
Broken installation.
Hey, guys, what are you doing? How you achieving it?
Amdgpu is not installed or not working, egl driver is not installed, RADV is not installed.
F means “full”, which includes the whole kitchen sink:
System
Machine
CPU
Graphics
Audio
Network
RAID
Drives
Partitions
Swap
Sensors
Misc info
For graphics issues, most of Faz is obfuscatory pollution (aka noise). Each of those 12 can be separately requested, as spelled out in man inxi, hence my previous request for GSaz (Graphics and System). “a” and “z” are modifiers: “a” means maximum detail for each section, as opposed to “x” (a bit more detail), “xx” (a bit more than a bit more) or “xxx” (lots more detail). “z” means omit sensitive data, such as serial numbers.
Thanks for all the tips.
I tried the IceWM desktop for a bit, but I couldn’t warm up to it.
Next I tried Wayland, just for comparison. So far my monitor hasn’t crashed while using Wayland. I wish it was the other way around, because Wayland has a number of strange effects that I find annoying.
I’ll stick with Wayland though, until I get some time to check the X11 session some more.
@Svyatko : this installation is from mid 2023, and has always been running fine, except for the recent HDMI problem. So I would not say that it’s broken.
I do see similar behavior. A monitor connected by HDMI on the computer side (the monitor only has DVI and an adapter is used) goes black for maybe 2 seconds. For me, the monitor comes back and everything is normal after that. This happens maybe 5 times a day and is therefore mildly annoying. I have seen this for a couple of month and had assumed that this might be a hardware problem with the monitor. No traces in the log files are left (dmesg).
I have been on ~ 5 tumbleweed snapshots currently on 20241111
Update: the problem seems to be gone!
This is what I did:
As reported already, I could run a Wayland session without the HDMI monitor shutting down. I used Wayland for a few days, no problem.
When I saw the suggestion from @sboehringer , I decided to retry the X11 session - which I much prefer - with the power management settings as suggested. Since then I have run X11 without problem.
Case closed for now. Thanks everyone for the help.