Application: akonadiserver (5.22.3 (22.12.3)) Qt Version: 5.15.9 Frameworks Version: 5.105.0 Operating System: Linux 6.4.6-100.fc37.x86_64 x86_64 Windowing System: Wayland Distribution: "Fedora release 37 (Thirty Seven)" DrKonqi: 5.27.4 [KCrashBackend] -- Information about the crash: I'd like to provide more information of what I was doing when the crash happened but unfortunately I wasn't doing something out of ordinary. Firefox, Telegram and VLC were open when suddenly Plasma crashed out of nowhere. The reporter is unsure if this crash is reproducible. -- Backtrace: Application: Akonadi Server (akonadiserver), signal: Segmentation fault [KCrash Handler] #4 0x000055792ceb029b in Akonadi::Server::AkonadiServer::connectionDisconnected() () #5 0x00007f17b10c8264 in QObject::event(QEvent*) () from /lib64/libQt5Core.so.5 #6 0x00007f17b109d5fb in QCoreApplication::notifyInternal2(QObject*, QEvent*) () from /lib64/libQt5Core.so.5 #7 0x00007f17b10a0984 in QCoreApplicationPrivate::sendPostedEvents(QObject*, int, QThreadData*) () from /lib64/libQt5Core.so.5 #8 0x00007f17b10eec27 in postEventSourceDispatch(_GSource*, int (*)(void*), void*) () from /lib64/libQt5Core.so.5 #9 0x00007f17af318c7f in g_main_context_dispatch () from /lib64/libglib-2.0.so.0 #10 0x00007f17af36f118 in g_main_context_iterate.constprop () from /lib64/libglib-2.0.so.0 #11 0x00007f17af315f00 in g_main_context_iteration () from /lib64/libglib-2.0.so.0 #12 0x00007f17b10ee71a in QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () from /lib64/libQt5Core.so.5 #13 0x00007f17b109c06a in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () from /lib64/libQt5Core.so.5 #14 0x00007f17b10a4132 in QCoreApplication::exec() () from /lib64/libQt5Core.so.5 #15 0x000055792cea6058 in main () [Inferior 1 (process 3393) detached] Reported using DrKonqi
Created attachment 160639 [details] New crash information added by DrKonqi akonadiserver (5.22.3 (22.12.3)) using Qt 5.15.9 Like OP, nothing unusual was happening. Only Firefox and VSCode windows were active. Nothing different from 90%+ of my KDE usage. -- Backtrace (Reduced): #4 0x0000557552b0429b in Akonadi::Server::AkonadiServer::connectionDisconnected() () #5 0x00007f98d7cc8264 in QObject::event(QEvent*) () from /lib64/libQt5Core.so.5 #6 0x00007f98d7c9d5fb in QCoreApplication::notifyInternal2(QObject*, QEvent*) () from /lib64/libQt5Core.so.5 #7 0x00007f98d7ca0984 in QCoreApplicationPrivate::sendPostedEvents(QObject*, int, QThreadData*) () from /lib64/libQt5Core.so.5 #8 0x00007f98d7ceec27 in postEventSourceDispatch(_GSource*, int (*)(void*), void*) () from /lib64/libQt5Core.so.5
In my case Telegram wasn't open. Firefox, Spotify, Thunderbird, and Proton Mail Bridge were.
(In reply to Alexia W from comment #2) > In my case Telegram wasn't open. Firefox, Spotify, Thunderbird, and Proton > Mail Bridge were. I'd forgotten I was also running VSCode. Upon launching it after the crash, the following error was displayed: > You're running in a KDE environment but the OS keyring is not available for encryption. Ensure you have kwallet running. Similar issues arose with Proton Mail Bridge and kwallet. Operating System: Fedora Linux 37 KDE Plasma Version: 5.27.4 KDE Frameworks Version: 5.105.0 Qt Version: 5.15.9 Kernel Version: 6.4.6-100.fc37.x86_64 (64-bit) Graphics Platform: Wayland
Similar versions to Alexia W, also running VS Code and Firefox, but didn't see that KWallet error on restart VSCode. Qt Version: 5.15.9 Frameworks Version: 5.105.0 Operating System: Linux 6.4.4-100.fc37.x86_64 x86_64 Windowing System: Wayland Distribution: Fedora Linux 37 (KDE Plasma) DrKonqi: 5.27.4 [KCrashBackend]
Created attachment 160651 [details] New crash information added by DrKonqi akonadiserver (5.22.3 (22.12.3)) using Qt 5.15.9 Crashed for me when Firefox and VSCode opened when I tried to switch from Firefox to VSCode -- Backtrace (Reduced): #4 0x000055b1ab5aa29b in Akonadi::Server::AkonadiServer::connectionDisconnected() () #5 0x00007f7552ac8264 in QObject::event(QEvent*) () from /lib64/libQt5Core.so.5 #6 0x00007f7552a9d5fb in QCoreApplication::notifyInternal2(QObject*, QEvent*) () from /lib64/libQt5Core.so.5 #7 0x00007f7552aa0984 in QCoreApplicationPrivate::sendPostedEvents(QObject*, int, QThreadData*) () from /lib64/libQt5Core.so.5 #8 0x00007f7552aeec27 in postEventSourceDispatch(_GSource*, int (*)(void*), void*) () from /lib64/libQt5Core.so.5
Created attachment 160655 [details] New crash information added by DrKonqi akonadiserver (5.22.3 (22.12.3)) using Qt 5.15.9 some terminator windows open (2 or 3? - not sure), 2 instances dolphin windows, vivaldi open; crash occured, when minimizing one of the dolphin instances; working on 4k screen connected using usb c; -- Backtrace (Reduced): #4 0x0000564e9291529b in Akonadi::Server::AkonadiServer::connectionDisconnected() () #5 0x00007ff853ac8264 in QObject::event(QEvent*) () from /lib64/libQt5Core.so.5 #6 0x00007ff853a9d5fb in QCoreApplication::notifyInternal2(QObject*, QEvent*) () from /lib64/libQt5Core.so.5 #7 0x00007ff853aa0984 in QCoreApplicationPrivate::sendPostedEvents(QObject*, int, QThreadData*) () from /lib64/libQt5Core.so.5 #8 0x00007ff853aeec27 in postEventSourceDispatch(_GSource*, int (*)(void*), void*) () from /lib64/libQt5Core.so.5
While the Akonadi report might not be useful, here's the logs from journalctl right around the crash, it seems Pipewire is also involved (it happens on 18:22:53): Jul 31 18:15:36 fedora pipewire[1442]: mod.client-node: 0x56053427c190: unknown peer 0x5605343d58d0 fd:101 Jul 31 18:16:13 fedora plasmashell[1569]: qt.qpa.wayland: Wayland does not support QWindow::requestActivate() Jul 31 18:16:13 fedora plasmashell[1569]: file:///usr/lib64/qt5/qml/org/kde/plasma/extras/PlaceholderMessage.qml:238:5: QML Heading: Binding loop detected for property "verticalAlignment" Jul 31 18:18:12 fedora pipewire[1442]: mod.client-node: 0x5605348449d0: unknown peer 0x56053436cae0 fd:52 Jul 31 18:18:42 fedora plasmashell[1569]: kpipewire_logging: PipeWire remote error: -2 unknown resource 3 op:4 Jul 31 18:18:42 fedora plasmashell[1569]: kpipewire_logging: PipeWire remote error: -2 unknown resource 3 op:4 Jul 31 18:18:42 fedora plasmashell[1569]: kpipewire_logging: PipeWire remote error: -2 unknown resource 3 op:4 Jul 31 18:18:42 fedora plasmashell[1569]: kpipewire_logging: PipeWire remote error: -2 unknown resource 3 op:4 Jul 31 18:18:42 fedora plasmashell[1569]: kpipewire_logging: PipeWire remote error: -2 unknown resource 3 op:4 Jul 31 18:18:42 fedora plasmashell[1569]: kpipewire_logging: PipeWire remote error: -2 unknown resource 3 op:7 Jul 31 18:18:54 fedora pipewire[1442]: mod.client-node: 0x5605343d52b0: unknown peer 0x560534370140 fd:52 Jul 31 18:20:36 fedora pipewire[1442]: mod.client-node: 0x56053427c230: unknown peer 0x560534363170 fd:97 Jul 31 18:20:36 fedora pipewire[1442]: mod.client-node: 0x56053427c230: unknown peer 0x560534363170 fd:97 Jul 31 18:21:06 fedora pipewire[1442]: mod.client-node: 0x5605343d52b0: unknown peer 0x56053436cae0 fd:92 Jul 31 18:21:07 fedora kwin_wayland[1437]: kwin_screencast: Dropping a screencast frame because the compositor is slow Jul 31 18:21:37 fedora pipewire[1442]: mod.client-node: 0x560534843cb0: unknown peer 0x56053436cae0 fd:52 Jul 31 18:22:44 fedora kwin_wayland[1437]: kwin_screencast: Dropping a screencast frame because the compositor is slow Jul 31 18:22:45 fedora kwin_wayland[1437]: kwin_screencast: Dropping a screencast frame because the compositor is slow Jul 31 18:22:47 fedora pipewire[1442]: mod.client-node: 0x5605343655b0: unknown peer 0x56053436cae0 fd:52 Jul 31 18:22:47 fedora pipewire[1442]: mod.client-node: 0x5605343650f0: unknown peer 0x560534370140 fd:91 Jul 31 18:22:47 fedora kwin_wayland[1437]: kwin_screencast: Dropping a screencast frame because the compositor is slow Jul 31 18:22:48 fedora pipewire[1442]: mod.client-node: 0x560534371fd0: unknown peer 0x560534363170 fd:129 Jul 31 18:22:53 fedora kwin_wayland[1437]: kwin_screencast: Dropping a screencast frame because the compositor is slow Jul 31 18:22:53 fedora audit[1437]: ANOM_ABEND auid=1000 uid=1000 gid=1000 ses=3 subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 pid=1437 comm="kwin_wayland" exe="/usr/bin/kwin_wayland" sig=11 res=1 Jul 31 18:22:53 fedora kernel: kwin_wayland[1437]: segfault at 7ff373353284 ip 00007ff3cbea10e1 sp 00007fff60b277f0 error 6 in libpipewire-0.3.so.0.376.0[7ff3cbe4d000+63000] likely on CPU 0 (core 0, socket 0) Jul 31 18:22:53 fedora kernel: Code: ca 01 00 48 8d 15 f7 b5 01 00 e8 8a d5 fa ff 59 5e e9 74 ff ff ff 0f 1f 00 f3 0f 1e fa 53 48 8b 46 28 48 89 fb 48 85 c0 74 05 <f0> 83 68 04 01 8b 46 24 a8 02 0f 85 07 01 00 00 8b 8b 18 10 00 00 Jul 31 18:22:53 fedora audit: BPF prog-id=82 op=LOAD Jul 31 18:22:53 fedora audit: BPF prog-id=83 op=LOAD Jul 31 18:22:53 fedora audit: BPF prog-id=84 op=LOAD Jul 31 18:22:53 fedora systemd[1]: Started systemd-coredump@1-12673-0.service - Process Core Dump (PID 12673/UID 0). ░░ Subject: A start job for unit systemd-coredump@1-12673-0.service has finished successfully
I've been able to reproduce this only using Firefox and Telegram desktop (and no audios were playing at the time). Seeing that the crash might be related to kwin_screencast, I assure you that there was no screencasting going on as far as I'm aware.
> when I tried to switch from Firefox to VSCode Same. A few seconds after I switched between Firefox to Telegram desktop. Plasma crashed and suddenly returned me to the desktop; all programs were closed, but no login was required. (Sorry for sending multiple messages in a row, I found no "edit message" button).
Re screencasts, I've just remembered that Firefox has been flaky with screensharing for a couple of days. Never had that problem before. Currently can't repo to get an exact error but the gist is "error: kwin_screencast can't create pipewire stream". Could be related..?
What you are seeing is kwin_wayland crashing. The crash report for Akonadi is misleading/not interesting. Could you please follow https://community.kde.org/Guidelines_and_HOWTOs/Debugging/How_to_create_useful_crash_reports#Retrieving_a_backtrace_using_coredumpctl to get a backtrace for kwin_wayland?
Hi, I've been running in the same issue since a couple of days. Here's the gdb backtrace from kwin_wayland from my side: (gdb) bt #0 0x00007fdde56a10e1 in pw_stream_queue_buffer () at /lib64/libpipewire-0.3.so.0 #1 0x000055ce3a037a5c in KWin::ScreenCastStream::enqueue() () #2 0x00007fdde5ad0fc6 in void doActivate<false>(QObject*, int, void**) () at /lib64/libQt5Core.so.5 #3 0x00007fdde5ad3833 in QSocketNotifier::activated(QSocketDescriptor, QSocketNotifier::Type, QSocketNotifier::QPrivateSignal) () at /lib64/libQt5Core.so.5 #4 0x00007fdde5ad4055 in QSocketNotifier::event(QEvent*) () at /lib64/libQt5Core.so.5 #5 0x00007fdde4faed72 in QApplicationPrivate::notify_helper(QObject*, QEvent*) () at /lib64/libQt5Widgets.so.5 #6 0x00007fdde5a9d618 in QCoreApplication::notifyInternal2(QObject*, QEvent*) () at /lib64/libQt5Core.so.5 #7 0x00007fdde5aebbbe in QEventDispatcherUNIXPrivate::activateSocketNotifiers() () at /lib64/libQt5Core.so.5 #8 0x00007fdde5aebffd in QEventDispatcherUNIX::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () at /lib64/libQt5Core.so.5 #9 0x000055ce3a0ee701 in QUnixEventDispatcherQPA::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () #10 0x00007fdde5a9c06a in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () at /lib64/libQt5Core.so.5 #11 0x00007fdde5aa4132 in QCoreApplication::exec() () at /lib64/libQt5Core.so.5 #12 0x000055ce3a00ed21 in main ()
Thanks. That looks like https://bugs.kde.org/show_bug.cgi?id=469949 Is anyone here using Plasma 5.27.6 where this is supposed to be fixed?
> Is anyone here using Plasma 5.27.6 where this is supposed to be fixed? I haven't experienced this problem after updating from 5.27.4 to 5.27.6. For Fedora 37 users, you need to upgrade to Fedora 38 to get Plasma 5.27.6; Or wait some time for 5.27.6 to land in Fedora 37 repos.
(In reply to Nicolas Fella from comment #13) > Thanks. That looks like https://bugs.kde.org/show_bug.cgi?id=469949 > > Is anyone here using Plasma 5.27.6 where this is supposed to be fixed? I can also confirm that moving to Fedora 38 with Plasma 5.27.6 apparently fixed the issue, as expected, for me. No crashes in days.
*** This bug has been marked as a duplicate of bug 469949 ***