Dear maintainers, k3b is crashing with with SIGBUS when trying to load libwidevinecdm.so from the local Firefox profile directory. A workaround is to set the environment variable QTWEBENGINE_CHROMIUM_FLAGS=--widevine-path=/does-not-exist This was reported as Debian bug #1091642 [1]. [1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1091642 SOFTWARE/OS VERSIONS Linux/KDE Plasma: Debian testing KDE Plasma Version: 6.2.4 KDE Frameworks Version: 6.8 Qt Version: 6.7.2 ADDITIONAL INFORMATION Logs : =============================================================================== QGuiApplication::setDesktopFileName: the specified desktop file name ends with .desktop. For compatibility reasons, the .desktop suffix will be removed. Please specify a desktop file name without .desktop suffix kf.coreaddons.kaboutdata: QGuiApplication::desktopFileName "org.kde.k3b" is out-of-sync with KAboutData::applicationData().desktopFileName "org.kde.k3b.desktop" QGuiApplication::setDesktopFileName: the specified desktop file name ends with .desktop. For compatibility reasons, the .desktop suffix will be removed. Please specify a desktop file name without .desktop suffix 21 21 kf.coreaddons.kaboutdata: QGuiApplication::desktopFileName "org.kde.k3b" is out-of-sync with KAboutData::applicationData().desktopFileName "org.kde.k3b.desktop" [124210:124210:1228/225345.267939:FATAL:zygote_host_impl_linux.cc(208)] Check failed: . : No such file or directory (2) fish: Job 1, 'k3b' terminated by signal SIGTRAP (Trace or breakpoint trap) =============================================================================== Stack : Thread 1 "k3b" received signal SIGTRAP, Trace/breakpoint trap. 0x00007fffed93a914 in logging::LogMessage::~LogMessage() () from /lib/x86_64-linux-gnu/libQt6WebEngineCore.so.6 (gdb) bt #0 0x00007fffed93a914 in logging::LogMessage::~LogMessage() () at /lib/x86_64-linux-gnu/libQt6WebEngineCore.so.6 #1 0x00007fffed93aa11 in logging::ErrnoLogMessage::~ErrnoLogMessage() () at /lib/x86_64-linux-gnu/libQt6WebEngineCore.so.6 #2 0x00007fffed93e0aa in logging::CheckError::~CheckError() () at /lib/x86_64-linux-gnu/libQt6WebEngineCore.so.6 #3 0x00007fffecb05277 in content::ZygoteHostImpl::LaunchZygote(base::CommandLine*, base::ScopedGeneric<int, base::internal::ScopedFDCloseTraits>*, std::vector<std::pair<int, int>, std::allocator<std::pair<int, int> > >) () at /lib/x86_64-linux-gnu/libQt6WebEngineCore.so.6 #4 0x00007fffed1f4b0a in content::(anonymous namespace)::LaunchZygoteHelper(base::CommandLine*, base::ScopedGeneric<int, base::internal::ScopedFDCloseTraits>*) () at /lib/x86_64-linux-gnu/libQt6WebEngineCore.so.6 #5 0x00007fffebf8ed11 in content::ZygoteCommunication::Init(base::OnceCallback<int (base::CommandLine*, base::ScopedGeneric<int, base::internal::ScopedFDCloseTraits>*)>) () at /lib/x86_64-linux-gnu/libQt6WebEngineCore.so.6 #6 0x00007fffebf2d785 in content::CreateGenericZygote(base::OnceCallback<int (base::CommandLine*, base::ScopedGeneric<int, base::internal::ScopedFDCloseTraits>*)>) () at /lib/x86_64-linux-gnu/libQt6WebEngineCore.so.6 #7 0x0000000000000030 in ??? () #8 0x0000000000000000 in ??? ()
This is clearly not a k3b problem, we're not debian nor firefox nor webengine/chromium there's nothing we can do here. I suggest you add more debug symbols so you get debug info on libQt6WebEngineCore and possibly report a bug against the debian webengine package since it seems what may be at fault here (or the Qt6 webengine code upstream).
Hmm, so k3b crashing on startup is not a k3b problem, oh well… I’ll try confirming with the original reporter if that happens with other webengine-enabled apps and get more symbols as you suggested. Thanks
I don't understand why you are upset, there's literally nothing we can do in k3b code to fix this problem (other than not using qtwebengine). Do it is not a k3b bug. It also "works for me" so it points to either a bug in qtwebengine, or in how debian compiled qtwebengine, again nothing we can fix.
For the record you were right that it has nothing to do with k3b specifically. The user had a corrupt widevine library in their browser config which caused the crash. It still looks very fragile to me and I’d be surprise if you cannot somehow attack that to inject random code into apps using qtwebengine.