Bug 385321

I am able to consistently recreate this as follows, although I am not sure if it is related to this bug or not.
I am using kscreenlocker 5.22.5 on Manjaro

----
Operating System: Manjaro Linux
KDE Plasma Version: 5.22.5
KDE Frameworks Version: 5.86.0
Qt Version: 5.15.2
Kernel Version: 5.14.7-2-MANJARO (64-bit)
Graphics Platform: Wayland
Processors: 4 × Intel® Core™ i5-6200U CPU @ 2.30GHz
Memory: 7.7 GiB of RAM
Graphics Processor: Mesa Intel® HD Graphics 520
----

I can consistently repeat is a follows:
1. While the session is unlocked, insert a usb sound card (I have an Avantree DG-80)
2. The system correctly switches to USB audio via pop-up notification
3. Close the laptop lid to force the laptop to sleep
4. While the laptop is sleeping, remove the usb sound card
5. Open the lid to wake from sleep
6. loginctl unlock-session is required

i've been seeing this for a while (Fedora 34-36) and put it down to a few things that I can't isolate easily.

Dell XPS13 9305 laptop (i5-1135g7) with built-in 4k screen on it's own works fine 100% reliably.
While the laptop is in sleep, if i connect or disconnect either the official Dell USB-C HDMI&USB dongle, or the Dell WD19TB dock with a 27" 2560x1440 screen then I'll see this issue maybe 80% of the time. being dongles/docks then i know there's multiple pieces of hardware being added/removed: display, audio, usb, etc etc.

if i do not change these connections while in sleep state then it appears reliable so it's like the screenlocker cannot handle a change in hardware while in a sleep state.

there also appears to be some sort of delay too; if i try to reproduce the issue with only a few seconds/minutes in sleep then the issue does not occur. if i leave it hours/overnight then the issue seems far more common.

I've no issues if i connect/disconnect the dock/dongle once the session is unlocked.

I'm hit by this as well. But what I think is even more concerning, why on earth do we need another screen/action to actually unlock the screen? Why not just enter the password and that's it - as it was before?

bye Mchi

I'm having this issue as well.

Operating System: Kubuntu 22.04
KDE Plasma Version: 5.24.6
KDE Frameworks Version: 5.95.0
Qt Version: 5.15.3
Kernel Version: 5.15.0-48-lowlatency (64-bit)
Graphics Platform: Wayland
Processors: 32 × AMD Ryzen 9 3950X 16-Core Processor
Memory: 31.3 GiB of RAM
Graphics Processor: AMD Radeon RX 5700 XT

9/15/22 10:02 AM	audit	ANOM_ABEND auid=1000 uid=1000 gid=1000 ses=3 subj=? pid=17710 comm="kscreenlocker_g" exe="/usr/lib/x86_64-linux-gnu/libexec/kscreenlocker_greet" sig=11 res=1
9/15/22 10:02 AM	kernel	kscreenlocker_g[17710]: segfault at 0 ip 0000000000000000 sp 00007ffcf74aff58 error 14 in kscreenlocker_greet[55ea0130e000+a000]
9/15/22 10:02 AM	kernel	Code: Unable to access opcode bytes at RIP 0xffffffffffffffd6.
9/15/22 10:02 AM	xdg-desktop-portal-kde	xdp-kde-wayland-integration: Adding output:
9/15/22 10:02 AM	xdg-desktop-portal-kde	xdp-kde-wayland-integration:     manufacturer:  "LG Electronics"
9/15/22 10:02 AM	xdg-desktop-portal-kde	xdp-kde-wayland-integration:     model:  "27GN950/007NTEP02148"
9/15/22 10:02 AM	xdg-desktop-portal-kde	xdp-kde-wayland-integration:     resolution:  QSize(3840, 2160)
9/15/22 10:02 AM	audit	MAC_TASK_CONTEXTS subj_apparmor=unconfined
9/15/22 10:02 AM	kernel	kscreenlocker_g[17806]: segfault at 40 ip 00007f334a031e94 sp 00007ffc1c671998 error 4 in libwayland-client.so.0.20.0[7f334a030000+6000]
9/15/22 10:02 AM	kernel	Code: ff ff ff ff eb d7 f3 0f 1e fa 48 89 77 30 c3 0f 1f 80 00 00 00 00 f3 0f 1e fa 48 8b 47 30 c3 0f 1f 80 00 00 00 00 f3 0f 1e fa <8b> 47 40 c3 0f 1f 84 00 00 00 00 00 f3 0f 1e fa 8b 47 10 c3 0f 1f

Steps to reproduce:
1. Lock screen or allow auto-lock due to time out
2. Allow screen to auto-off or manually turn it off
3. Displays the following:
"The screen locker is broken and unlocking is not possible anymore. In order to unlock switch to a virtual terminal(Ctrl+Alt+F2), login and execute the command loginctl unlock-session 2. Afterwards switch back to the running session (Ctrl+Alt+F1)"

Session successfully recovers if steps are followed on the virtual terminal. Reproducible every time.

can confirm its still happening on 5.25.5 (current KDE Fedora36 version) as i got the issue over the last few days. When it happens again i'll add some further info as it's still appears randomly intermittent.

since my last comment in august, I've since sold the XPS and now on loan Dell 5410, i5-8gen with same symptoms. I've a i5-12th gen laptop due next week so i expect the issue to occur on that too. They're all using the intel iGPU of varying ages. My screen resolutions still differ between the laptop and external screen (i think its been established it occurs on mis-matched resolutions?)

i can see 5.26.2 is currently targeted for F37 which is due on 15th Nov so i'll be able to test that once released.

I've since upgraded packages to the bellow specifications, and it is still occuring; however it is now somewhat up to chance if it occurs instead of reliably replicable like my last report.

Operating System: Kubuntu 22.04
KDE Plasma Version: 5.25.5
KDE Frameworks Version: 5.98.0
Qt Version: 5.15.3
Kernel Version: 5.15.0-53-lowlatency (64-bit)
Graphics Platform: Wayland

(In reply to veehexx from comment #13)
> can confirm its still happening on 5.25.5 (current KDE Fedora36 version) as
> i got the issue over the last few days. When it happens again i'll add some
> further info as it's still appears randomly intermittent.
> 
> since my last comment in august, I've since sold the XPS and now on loan
> Dell 5410, i5-8gen with same symptoms. I've a i5-12th gen laptop due next
> week so i expect the issue to occur on that too. They're all using the intel
> iGPU of varying ages. My screen resolutions still differ between the laptop
> and external screen (i think its been established it occurs on mis-matched
> resolutions?)
> 
> i can see 5.26.2 is currently targeted for F37 which is due on 15th Nov so
> i'll be able to test that once released.

my new laptop turned up last week and been running Fedora 36, and last night upgraded to F37. i've not had any issues with this laptop other than the second 'unlock' button screen. no manual loginctl method required. i'm not sure what the differences are to cause this. The laptop and another USB-C dock with higher PD delivery are the only things that have changed. i still have mismatched screen resolutions.
I'll update if i experience the issue. I'll also try to reproduce with the old USB-C dock to see if that is part of the issue sometime over the next few days.

A lot has changed since the original bug report; are you still able to reproduce this kde@mnetwork.co.kr?

If not, let's use it for veehexx@zoho.com's issue--while hopefully is also fixed!

In general for these kinds of issues it's always good to file a new bug report, since an issue with the same symptoms can be caused by wildly different root causes.

(In reply to Nate Graham from comment #16)
> In general for these kinds of issues it's always good to file a new bug
> report, since an issue with the same symptoms can be caused by wildly
> different root causes.

I'm happy to create my own bug report and link back. Would you like me to do that?
I've had the loginctl finally happen to me tonight.

> ...
> [Fri Nov 18 18:06:26 2022] wlp166s0: authenticate with e8:48:b8:9b:4a:45
> [Fri Nov 18 18:06:26 2022] wlp166s0: send auth to e8:48:b8:9b:4a:45 (try 1/3)
> [Fri Nov 18 18:06:26 2022] wlp166s0: authenticate with e8:48:b8:9b:4a:45
> [Fri Nov 18 18:06:26 2022] wlp166s0: send auth to e8:48:b8:9b:4a:45 (try 1/3)
> [Fri Nov 18 18:06:26 2022] wlp166s0: authenticated
> [Fri Nov 18 18:06:26 2022] wlp166s0: associate with e8:48:b8:9b:4a:45 (try 1/3)
> [Fri Nov 18 18:06:26 2022] wlp166s0: RX AssocResp from e8:48:b8:9b:4a:45 (capab=0x511 status=0 aid=2)
> [Fri Nov 18 18:06:26 2022] wlp166s0: associated
> [Fri Nov 18 18:06:26 2022] wlp166s0: Limiting TX power to 27 (30 - 3) dBm as advertised by e8:48:b8:9b:4a:45
> [Fri Nov 18 18:06:26 2022] IPv6: ADDRCONF(NETDEV_CHANGE): wlp166s0: link becomes ready
> [Fri Nov 18 18:06:28 2022] wlp166s0: deauthenticating from e8:48:b8:9b:4a:45 by local choice (Reason: 3=DEAUTH_LEAVING)
> [Fri Nov 18 18:06:29 2022] show_signal_msg: 63 callbacks suppressed
> [Fri Nov 18 18:06:29 2022] kscreenlocker_g[3741]: segfault at 18 ip 00007f9a0f90bda6 sp 00007ffdab661660 error 4 in libLayerShellQtInterface.so.5.26.3[7f9a0f90a000+3000]
> [Fri Nov 18 18:06:29 2022] Code: 8d 64 24 38 48 89 44 24 28 48 89 ee 4c 89 e7 e8 10 e8 ff ff 48 89 ef e8 08 e9 ff ff 4c 89 e6 4c 89 ef 89 44 24 24 49 8b 46 38 <4c> 8b 48 18 4c 89 4c 24 18 e8 cc e9 ff ff 48 8b 44 24 40 4d 8b 7f
> [Fri Nov 18 18:06:29 2022] PM: suspend entry (s2idle)
> [Fri Nov 18 18:06:29 2022] Filesystems sync: 0.105 seconds
> [Fri Nov 18 18:06:29 2022] Freezing user space processes ... (elapsed 0.002 seconds) done.
> [Fri Nov 18 18:06:29 2022] OOM killer disabled.
> [Fri Nov 18 18:06:29 2022] Freezing remaining freezable tasks ... (elapsed 0.001 seconds) done.
> [Fri Nov 18 18:06:29 2022] printk: Suspending console(s) (use no_console_suspend to debug)
> [Fri Nov 18 18:06:30 2022] PM: suspend devices took 0.593 seconds
> [Fri Nov 18 18:06:30 2022] ACPI: EC: interrupt blocked
> ...

OS: Fedora 37 KDE spin
KDE plasma version: 5.26.3
Frameworks version: 5.100.0
QT version: 5.15:6
kernel: 6.0.8-300.fc37
Graphics platform: Wayland

The exact process I've done tonight is disconnect laptop from charger/usb dock, update the entire OS via KDE Discover(i noticed the general OS updates and a handful of KDE flatpak updates) and reboot to apply the various updates. Login and fairly quickly press powerbutton to sleep. I come back about an hour later to wake laptop up and thats when i'm presented by the loginctl screen.

Based on the above dmesg output, kscreenlocker_g crashed as the laptop went into sleep, and not on the way out. First time I've had an issue since owning this laptop (about 2 weeks).
Laptop is a Framework i5-1240P model.

Yes, it would be helpful if you could create a separate bug report for it.

done. https://bugs.kde.org/show_bug.cgi?id=462010

Thank you!

Hello!
I'm newbe here, but I have something to share on reproduction and a life hack to bypass the error from the subject.

I am using "Rosa Linux R12.3" with:
KDE Plama = 5.26.3
KDE Frameworks = 5.100.0
Qt = 5.15.5

I have a habit of blocking the screen when leaving the workplace using the "Win + L" combination.

When a screen lock error occurs, I get the following in the log:

ноя 22 15:07:08 r12 kscreenlocker_greet[2346]: Qt: Session management error: networkIdsList argument is NULL
ноя 22 15:07:08 r12 kscreenlocker_greet[2346]: kf.kirigami: Failed to find a Kirigami platform plugin
ноя 22 15:07:08 r12 ksmserver[1704]: error in client communication (pid 1704)
ноя 22 15:07:08 r12 ksmserver[2346]: wl_display@1: error 0: invalid object 0
ноя 22 15:07:08 r12 audit[2346]: ANOM_ABEND auid=500 uid=500 gid=500 ses=3 pid=2346 comm="kscreenlocker_g" exe="/usr/lib64/libexec/kscreenlocker_greet" sig=11 res=1
ноя 22 15:07:08 r12 kernel: kscreenlocker_g[2346]: segfault at 40 ip 00007fb3d7c67700 sp 00007ffd1a86d308 error 4 in libwayland-client.so.0.21.0[7fb3d7c65000+6000]
ноя 22 15:07:08 r12 kernel: Code: 00 e9 24 dc ff ff 0f 1f 40 00 48 89 77 30 c3 66 66 2e 0f 1f 84 00 00 00 00 00 48 8b 47 30 c3 66 66 2e 0f 1f 84 00 00 00 00 00 <8b> 47 40 c3 66 66 2e 0f 1f 84 00 00 00 00 00 90 8b 47 10 c3 66 66

When the blocker works correctly, the following is in the log:

ноя 22 15:14:43 r12 kscreenlocker_greet[2646]: Qt: Session management error: networkIdsList argument is NULL
ноя 22 15:14:44 r12 kscreenlocker_greet[2646]: kf.kirigami: Failed to find a Kirigami platform plugin
ноя 22 15:14:45 r12 kscreenlocker_greet[2646]: qt.virtualkeyboard.hunspell: Hunspell dictionary is missing for "ru_RU" . Search paths ("/usr/share/qt5/qtvirtualkeyboard/hunspell", "/usr/share/hunspell", "/usr/share/myspell/dicts")
ноя 22 15:14:48 r12 kscreenlocker_greet[2646]: pam_succeed_if(kde:auth): requirement "user ingroup nopasswdlogin" not met by user "det"
ноя 22 15:14:50 r12 audit[2664]: USER_AUTH pid=2664 uid=500 auid=500 ses=3 msg='op=PAM:unix_chkpwd acct="det" exe="/sbin/unix_chkpwd" hostname=? addr=? terminal=? res=success'

А life hack to bypass the error from the subject:
I will add that if you keep the console window open and "Dolphin" or another window from "KDE", for example "Settings", then the screen lock error does not reproduce.

In the window manager, configure the movement of two windows to the second desktop, for example, the console window and system settings;
Add the same applications to autorun;
When you log in to your account, go to the first desktop and work.

I am also ready to provide any additional information on the system and its environment, but I don’t know yet what kind of information you might need.

This sounds like different issue compared to the one in the OP; can you submit a new bug report for it? Thanks!