Application: krunner (5.5.4) Qt Version: 5.5.1 Operating System: Linux 4.3.3-303.fc23.x86_64 x86_64 Distribution: "Fedora release 23 (Twenty Three)" -- Information about the crash: - Unusual behavior I noticed: Display froze up when doing a cleanup of cache files with Bleachbit 1.10 as user (not admin) The crash can be reproduced every time. -- Backtrace: Application: krunner (krunner), signal: Bus error Using host libthread_db library "/lib64/libthread_db.so.1". [Current thread is 1 (Thread 0x7f7f82ec4940 (LWP 2543))] Thread 3 (Thread 0x7f7f63dc4700 (LWP 2545)): #0 0x00007f7f7cf5cfdd in poll () at /lib64/libc.so.6 #1 0x00007f7f7bdb3272 in _xcb_conn_wait () at /lib64/libxcb.so.1 #2 0x00007f7f7bdb4ee7 in xcb_wait_for_event () at /lib64/libxcb.so.1 #3 0x00007f7f65f5cda9 in QXcbEventReader::run() () at /lib64/libQt5XcbQpa.so.5 #4 0x00007f7f7db653de in QThreadPrivate::start(void*) () at /lib64/libQt5Core.so.5 #5 0x00007f7f79edd60a in start_thread () at /lib64/libpthread.so.0 #6 0x00007f7f7cf68a4d in clone () at /lib64/libc.so.6 Thread 2 (Thread 0x7f7f5b78d700 (LWP 2553)): #0 0x00007f7f7cf5cfdd in poll () at /lib64/libc.so.6 #1 0x00007f7f783a516c in g_main_context_iterate.isra () at /lib64/libglib-2.0.so.0 #2 0x00007f7f783a527c in g_main_context_iteration () at /lib64/libglib-2.0.so.0 #3 0x00007f7f7dd9cecb in QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () at /lib64/libQt5Core.so.5 #4 0x00007f7f7dd43eca in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () at /lib64/libQt5Core.so.5 #5 0x00007f7f7db60434 in QThread::exec() () at /lib64/libQt5Core.so.5 #6 0x00007f7f816e59b5 in QQmlThreadPrivate::run() () at /lib64/libQt5Qml.so.5 #7 0x00007f7f7db653de in QThreadPrivate::start(void*) () at /lib64/libQt5Core.so.5 #8 0x00007f7f79edd60a in start_thread () at /lib64/libpthread.so.0 #9 0x00007f7f7cf68a4d in clone () at /lib64/libc.so.6 Thread 1 (Thread 0x7f7f82ec4940 (LWP 2543)): [KCrash Handler] #5 0x00007f7f7e94251e in KSharedDataCache::clear() () at /lib64/libKF5CoreAddons.so.5 #6 0x00007f7f826fd13a in Plasma::ThemePrivate::discardCache(QFlags<Plasma::CacheType>) () at /lib64/libKF5Plasma.so.5 #7 0x00007f7f827066bb in Plasma::ThemePrivate::qt_static_metacall(QObject*, QMetaObject::Call, int, void**) () at /lib64/libKF5Plasma.so.5 #8 0x00007f7f7dd7529a in QMetaObject::activate(QObject*, int, int, void**) () at /lib64/libQt5Core.so.5 #9 0x00007f7f7dd81b48 in QTimer::timerEvent(QTimerEvent*) () at /lib64/libQt5Core.so.5 #10 0x00007f7f7dd76113 in QObject::event(QEvent*) () at /lib64/libQt5Core.so.5 #11 0x00007f7f7ecc441c in QApplicationPrivate::notify_helper(QObject*, QEvent*) () at /lib64/libQt5Widgets.so.5 #12 0x00007f7f7ecc98e6 in QApplication::notify(QObject*, QEvent*) () at /lib64/libQt5Widgets.so.5 #13 0x00007f7f7dd4673b in QCoreApplication::notifyInternal(QObject*, QEvent*) () at /lib64/libQt5Core.so.5 #14 0x00007f7f7dd9bccd in QTimerInfoList::activateTimers() () at /lib64/libQt5Core.so.5 #15 0x00007f7f7dd9c1d1 in timerSourceDispatch(_GSource*, int (*)(void*), void*) () at /lib64/libQt5Core.so.5 #16 0x00007f7f783a4e3a in g_main_context_dispatch () at /lib64/libglib-2.0.so.0 #17 0x00007f7f783a51d0 in g_main_context_iterate.isra () at /lib64/libglib-2.0.so.0 #18 0x00007f7f783a527c in g_main_context_iteration () at /lib64/libglib-2.0.so.0 #19 0x00007f7f7dd9ceaf in QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () at /lib64/libQt5Core.so.5 #20 0x00007f7f7dd43eca in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () at /lib64/libQt5Core.so.5 #21 0x00007f7f7dd4bfac in QCoreApplication::exec() () at /lib64/libQt5Core.so.5 #22 0x00005646d820fed7 in main () Reported using DrKonqi
seems some symbols for KSharedDataCache are missing
possible thread unsafety of kshareddatacache?
Bleachbit overwrites the cache instead of simply unlinking it, which is what causes the crash. There's not much I can do here because the cache is persistent so it must save its accounting data within the file itself, which Bleachbit then corrupts, leading to crashes almost instantly. This type of bug has been reported before, my understanding was that the Bleachbit devs had configured the software to leave things like /var/tmp/kdecache* alone by default but I'm not sure (and I'm sure it's still possible for user to override). See Martin's blog post from nearly 4 years ago now to see how long this has been going on https://blog.martin-graesslin.com/blog/2012/07/do-not-use-bleachbit/
Created attachment 97348 [details] attachment-19889-0.html Thank you much for the information about Bleachbit's causation of this crash. I will cease using it. Your reply is very much appreciated. Joe G On Feb 19, 2016 9:39 PM, "Michael Pyne via KDE Bugzilla" < bugzilla_noreply@kde.org> wrote: > https://bugs.kde.org/show_bug.cgi?id=358887 > > Michael Pyne <mpyne@kde.org> changed: > > What |Removed |Added > > ---------------------------------------------------------------------------- > Status|NEEDSINFO |RESOLVED > Resolution|BACKTRACE |INVALID > > --- Comment #3 from Michael Pyne <mpyne@kde.org> --- > Bleachbit overwrites the cache instead of simply unlinking it, which is > what > causes the crash. There's not much I can do here because the cache is > persistent so it must save its accounting data within the file itself, > which > Bleachbit then corrupts, leading to crashes almost instantly. > > This type of bug has been reported before, my understanding was that the > Bleachbit devs had configured the software to leave things like > /var/tmp/kdecache* alone by default but I'm not sure (and I'm sure it's > still > possible for user to override). See Martin's blog post from nearly 4 years > ago > now to see how long this has been going on > https://blog.martin-graesslin.com/blog/2012/07/do-not-use-bleachbit/ > > -- > You are receiving this mail because: > You reported the bug. >