Bug 493374

Summary: Black window with only cursor after sleep
Product: [Plasma] plasmashell Reporter: nilskemail+kde
Component: generic-crashAssignee: Plasma Bugs List <plasma-bugs-null>
Status: RESOLVED DUPLICATE    
Severity: normal CC: nate
Priority: NOR    
Version First Reported In: 6.1.4   
Target Milestone: 1.0   
Platform: Fedora RPMs   
OS: Linux   
Latest Commit: Version Fixed/Implemented In:
Sentry Crash Report:
Attachments: System log while reproducing the bug
User log while reproducing the bug

Description nilskemail+kde 2024-09-19 17:30:25 UTC
SUMMARY
When putting the device to sleep and shortly afterwards closing the lid, the subsequent waking only results in a black screen with a cursor. Closing the lid a second time and opening it again results plasmashell (or kwin, or sddm?) terminating and starting a completely new session

STEPS TO REPRODUCE
1. Invoke sleep action
2. Close lid
3. Reopen lid

OBSERVED RESULT
A black screen with only a cursor appears (the cursor is not frozen, and shake to embiggen works)

EXPECTED RESULT
A login screen should appear

SOFTWARE/OS VERSIONS
Operating System: Fedora Linux 40
KDE Plasma Version: 6.1.4
KDE Frameworks Version: 6.5.0
Qt Version: 6.7.2
Kernel Version: 6.10.7-200.fc40.x86_64 (64-bit)
Graphics Platform: Wayland
Processors: 16 × AMD Ryzen 7 7840U w/ Radeon 780M Graphics
Memory: 30.7 GiB of RAM
Graphics Processor: AMD Radeon 780M

ADDITIONAL INFORMATION
This is not completely deterministic but happens more often than not
Comment 1 Nate Graham 2024-09-19 19:15:35 UTC
This seems bad. When you do it a second time, it sounds like KWin is crashing. Can you get a backtrace of the kwin_wayland crash? See https://community.kde.org/Guidelines_and_HOWTOs/Debugging/How_to_create_useful_crash_reports#Retrieving_a_backtrace_using_coredumpctl. Hopefully that will give us a clue regarding what's happening here. Thanks a lot!
Comment 2 nilskemail+kde 2024-09-20 15:00:19 UTC
(In reply to Nate Graham from comment #1)
> This seems bad. When you do it a second time, it sounds like KWin is
> crashing. Can you get a backtrace of the kwin_wayland crash?

There is nothing related to that crash listed by coredumpctl. Not kwin or anything else
Comment 3 Nate Graham 2024-09-20 17:52:39 UTC
Could be that KWin is exiting with a Wayland protocol error, then. Since that's a secondary effect, possibly triggered by the initial issue, let's dig into the initial issue. It seems that plasmashell has also exited rather than crashing. The next time this happens, can you use KRunner (Alt+Space) to open Konsole and then use Konsole to run `journalctl --reverse` and see if there are any suspicious-looking messages from plasmashell? Just copy and paste the last minute's worth of messages here. Thanks!
Comment 4 nilskemail+kde 2024-09-20 20:04:03 UTC
Created attachment 173921 [details]
System log while reproducing the bug
Comment 5 nilskemail+kde 2024-09-20 20:04:31 UTC
Created attachment 173922 [details]
User log while reproducing the bug
Comment 6 nilskemail+kde 2024-09-20 20:07:18 UTC
I reproduced the bug while running journalctl --follow (with and without --user) and piping the output into a file.
Comment 7 nilskemail+kde 2024-09-20 23:01:27 UTC
I also just noticed that when I activate the lock screen using Meta+L, move the cursor to unlock it (I have an empty password), and lock the screen again using Meta+L, I can achieve a similar situation. The screen is also black and the cursor is visible but moving it does not unlock the PC. However, unlike the original scenario after a wake-up from sleep, I can put any finger on the fingerprint reader (also unregistered ones) which unlocks the PC and shows my plasma session again. This is not possible with the scenario when waking up from sleep after closing the laptop lid though...

Even though these are a bit different, I feel like they might be related or share the same cause. It is weird that one can be resolved by starting a scan with the fingerprint sensor (even a failing one) while the other one is not affected by it but also the fact that this, as well as the reproduction in the video, happens at the second time only seems like a very odd coincidence - and, well: There are no coincidences. I am not sure how I encountered this initially but so far all my reproductions with sleeping/lid closing also happened on the second try only IIRC...
Comment 8 Nate Graham 2024-09-24 15:30:30 UTC
Aha, this is Bug 493572.

*** This bug has been marked as a duplicate of bug 493572 ***