Version: 1.9.50 (using 4.00.80 (KDE 4.0.80 >= 20080104), Gentoo) Compiler: i686-pc-linux-gnu-gcc OS: Linux (i686) release 2.6.22-gentoo-r9 Today when I was moving mail around in Kontact in order to try to reproduce bug #156291 KMail all of a sudden (seemingly randomly) crashed with the following backtrace: Using host libthread_db library "/lib/libthread_db.so.1". [Thread debugging using libthread_db enabled] [New Thread 0xb53fe6d0 (LWP 26320)] [KCrash handler] #6 KMail::ActionScheduler::processMessage (this=0x84e3630) at /var/tmp/portage/kde-base/kdepim-9999.4/work/kdepim-9999.4/kmail/actionscheduler.cpp:533 #7 0xb3569181 in KMail::ActionScheduler::qt_metacall (this=0x84e3630, _c=QMetaObject::InvokeMetaMethod, _id=18, _a=0xbf9e8434) at /var/tmp/portage/kde-base/kdepim-9999.4/work/kdepim_build/kmail/actionscheduler.moc:118 #8 0xb7f441f4 in QMetaObject::activate (sender=0x84deeb8, from_signal_index=4, to_signal_index=4, argv=<value optimized out>) at kernel/qobject.cpp:3081 #9 0xb7f44d94 in QMetaObject::activate (sender=0x84deeb8, m=0xb7fe39e4, local_signal_index=0, argv=0x0) at kernel/qobject.cpp:3140 #10 0xb7f65083 in QTimer::timeout (this=0x84deeb8) at .moc/debug-shared/moc_qtimer.cpp:125 #11 0xb7f4da2e in QTimer::timerEvent (this=0x84deeb8, e=0xbf9e8968) at kernel/qtimer.cpp:259 #12 0xb7f42269 in QObject::event (this=0x84deeb8, e=0x8065138) at kernel/qobject.cpp:1086 #13 0xb6e04e8a in QApplicationPrivate::notify_helper (this=0x8060068, receiver=0x84deeb8, e=0xbf9e8968) at kernel/qapplication.cpp:3556 #14 0xb6e0677a in QApplication::notify (this=0xbf9e8bf0, receiver=0x84deeb8, e=0xbf9e8968) at kernel/qapplication.cpp:3115 #15 0xb7ab6fb3 in KApplication::notify (this=0xbf9e8bf0, receiver=0x84deeb8, event=0xbf9e8968) at /var/tmp/portage/kde-base/kdelibs-9999.4/work/kdelibs-9999.4/kdeui/kernel/kapplication.cpp:314 #16 0xb7f30d7b in QCoreApplication::notifyInternal (this=0xbf9e8bf0, receiver=0x84deeb8, event=0xbf9e8968) at kernel/qcoreapplication.cpp:530 #17 0xb7f59efb in QTimerInfoList::activateTimers (this=0x80607e4) at kernel/qcoreapplication.h:200 #18 0xb7f59fd1 in QEventDispatcherUNIX::activateTimers (this=0x805f170) at kernel/qeventdispatcher_unix.cpp:828 #19 0xb7f5a9ab in QEventDispatcherUNIX::processEvents (this=0x805f170, flags=@0xbf9e8ae8) at kernel/qeventdispatcher_unix.cpp:890 #20 0xb6e8ecce in QEventDispatcherX11::processEvents (this=0x805f170, flags=@0xbf9e8b14) at kernel/qeventdispatcher_x11.cpp:145 #21 0xb7f30191 in QEventLoop::processEvents (this=0xbf9e8b80, flags=@0xbf9e8b48) at kernel/qeventloop.cpp:140 #22 0xb7f3029a in QEventLoop::exec (this=0xbf9e8b80, flags=@0xbf9e8b88) at kernel/qeventloop.cpp:186 #23 0xb7f32626 in QCoreApplication::exec () at kernel/qcoreapplication.cpp:759 #24 0xb6e04487 in QApplication::exec () at kernel/qapplication.cpp:3053 #25 0x0804b3f5 in main (argc=1, argv=0xbf9e8d84) at /var/tmp/portage/kde-base/kdepim-9999.4/work/kdepim-9999.4/kontact/src/main.cpp:162 #26 0xb577ffdc in __libc_start_main (main=0x804ab60 <main>, argc=1, ubp_av=0xbf9e8d84, init=0x804c180 <__libc_csu_init>, fini=0x804c170 <__libc_csu_fini>, rtld_fini=0xb7ff4100 <_dl_fini>, stack_end=0xbf9e8d7c) at libc-start.c:229 #27 0x0804a791 in _start () #0 0xffffe410 in __kernel_vsyscall ()
Is it possible that this bug is related to bug #156289? The backtraces do not look too different, and after an update that fixed bug #156289 I can neither reproduce bug #156289 nor this bug? (I'll mark this bug as resolved for that reason)
Sorry, posted the comment above on the wrong bug :/
SVN commit 818122 by tmcguire: Don't crash in a debug function. This might also fix bug 156319, although the backtrace misses the last function it it. CCBUG: 156319 M +0 -1 actionscheduler.cpp WebSVN link: http://websvn.kde.org/?view=rev&revision=818122
Did that patch fix the issue for you Oliver?
Marking as NEEDSINFO until Oliver replies. So far, nobody else reported such issue.
Ups, sorry for the late reply. Apparently I completely missed the first message from Michael... At any rate: The problem with confirming that the fix fixed the problem is that I was never able to reliably reproduce this crash at will to begin with. So I'd suggest we close this as (supposedly) fixed, and if I encounter a crash with a similar backtrace to the original one in the future (so far kmail seems to have stabilized quite a bit), I just reopen this bugreport...