Application that crashed: kontact Version of the application: 4.3.0 rc1 KDE Version: 4.2.95 (KDE 4.2.95 (KDE 4.3 RC1)) Qt Version: 4.5.2 Operating System: Linux 2.6.29-gentoo-r5 x86_64 What I was doing when the application crashed: One thing I noticed when exiting, in fact it was the reason I was exiting, is that there was still some scan of a remote mailbox going on, and had been going on for hours. It might have been IMAP being stuck, I don't know, but I figured I'd exit and restart just to reset that. And then I got this crash. -- Backtrace: Application: Kontact (kontact), signal: Segmentation fault [KCrash Handler] #5 0x00007ff2e354e740 in QListData::shared_null () from /usr/lib64/qt4/libQtCore.so.4 #6 0x00007ff2e57b03e6 in qDeleteAll<QList<KIO::Slave*>::const_iterator> (begin={i = 0x7ffff1522b10}, end={i = 0x7ffff1522b00}) at /usr/include/qt4/QtCore/qalgorithms.h:350 #7 0x00007ff2e57b0db2 in qDeleteAll<QList<KIO::Slave*> > (c=@0x7ffff1522b40) at /usr/include/qt4/QtCore/qalgorithms.h:358 #8 0x00007ff2e57b163d in ~ProtocolInfo (this=0xca69f0) at /var/tmp/portage/kde-base/kdelibs-4.2.95/work/kdelibs-4.2.95/kio/kio/scheduler.cpp:171 #9 0x00007ff2e57b16d0 in qDeleteAll<QHash<QString, KIO::SchedulerPrivate::ProtocolInfo*>::const_iterator> (begin={i = 0x14bd4e0}, end={i = 0x9f69d0}) at /usr/include/qt4/QtCore/qalgorithms.h:350 #10 0x00007ff2e57b194a in ~SchedulerPrivate (this=0xb82160) at /var/tmp/portage/kde-base/kdelibs-4.2.95/work/kdelibs-4.2.95/kio/kio/scheduler.cpp:103 #11 0x00007ff2e57abef2 in destroy () at /var/tmp/portage/kde-base/kdelibs-4.2.95/work/kdelibs-4.2.95/kio/kio/scheduler.cpp:209 #12 0x00007ff2e2400dad in exit () from /lib/libc.so.6 #13 0x00007ff2e23ea5cd in __libc_start_main () from /lib/libc.so.6 #14 0x0000000000403289 in _start () This bug may be a duplicate of or related to bug 195896 Reported using DrKonqi
Indeed, it looks related to bug 195896. Do you think you would be able to reproduce the crash situation at will? Thanks
*** Bug 195896 has been marked as a duplicate of this bug. ***
*** Bug 192858 has been marked as a duplicate of this bug. ***
*** Bug 200389 has been marked as a duplicate of this bug. ***
*** Bug 202735 has been marked as a duplicate of this bug. ***
*** Bug 203433 has been marked as a duplicate of this bug. ***
*** Bug 203717 has been marked as a duplicate of this bug. ***
*** Bug 204069 has been marked as a duplicate of this bug. ***
Happened on Gentoo as well, KDE from 4.3 branch (as of 22.08.2009), Qt-4.5.1, using IMAP. Application: KMail (kmail), signal: Segmentation fault [KCrash Handler] #5 0x00007ffa0abd7720 in QListData::shared_null () from /usr/lib64/qt4/libQtCore.so.4 #6 0x00007ffa04ac073b in qDeleteAll<QList<KIO::Slave*>::const_iterator> (begin=..., end=...) at /usr/include/qt4/QtCore/qalgorithms.h:350 #7 0x00007ffa04ac1119 in qDeleteAll<SlaveList> (c=...) at /usr/include/qt4/QtCore/qalgorithms.h:358 #8 0x00007ffa04ac1d56 in KIO::SchedulerPrivate::ProtocolInfo::~ProtocolInfo (this=0x171c370, __in_chrg=<value optimized out>) at /var/tmp/portage/kde-base/kdelibs-4.3.9999/work/kdelibs-4.3.9999/kio/kio/scheduler.cpp:171 #9 0x00007ffa04ac1ddd in qDeleteAll<QHash<QString, KIO::SchedulerPrivate::ProtocolInfo*>::const_iterator> (begin=..., end=...) at /usr/include/qt4/QtCore/qalgorithms.h:350 #10 0x00007ffa04ac1e2f in qDeleteAll<KIO::SchedulerPrivate::ProtocolInfoDict> (c=...) at /usr/include/qt4/QtCore/qalgorithms.h:358 #11 0x00007ffa04ac2068 in KIO::SchedulerPrivate::~SchedulerPrivate (this=0x1283500, __in_chrg=<value optimized out>) at /var/tmp/portage/kde-base/kdelibs-4.3.9999/work/kdelibs-4.3.9999/kio/kio/scheduler.cpp:103 #12 0x00007ffa04abc511 in <anonymous struct>::destroy(void) () at /var/tmp/portage/kde-base/kdelibs-4.3.9999/work/kdelibs-4.3.9999/kio/kio/scheduler.cpp:209 #13 0x00007ffa04a15571 in KCleanUpGlobalStatic::~KCleanUpGlobalStatic (this=0x7ffa04df7c58, __in_chrg=<value optimized out>) at /var/tmp/portage/kde-base/kdelibs-4.3.9999/work/kdelibs-4.3.9999/kdecore/kernel/kglobal.h:62 #14 0x00007ffa09f926fd in exit () from /lib/libc.so.6 #15 0x00007ffa09f7a5cd in __libc_start_main () from /lib/libc.so.6 #16 0x00000000004024d9 in _start ()
*** Bug 204898 has been marked as a duplicate of this bug. ***
*** Bug 205021 has been marked as a duplicate of this bug. ***
A bit off topic, but with nine (9) duplicates, it's slightly embarrassing to consider this as UNCONFIRMED.
Bug 205083 seems to be the same as this but in Konqueror, so this could be a KIO issue not related to Kontact. @David: could you confirm this ? Thanks
I have Gentoo and kde-4.3.0. Kmail crashes on exit while opening mail using imap. Emails from local folders open too fast, so I cannot close kmail while opening them. Konqueror works fine for me.
@Dario Indeed judging from backtrace one is duplicate of the other. Also still reproducible on 4.3 branch with qt-4.5.2
*** Bug 205777 has been marked as a duplicate of this bug. ***
*** Bug 206392 has been marked as a duplicate of this bug. ***
*** Bug 206509 has been marked as a duplicate of this bug. ***
*** Bug 207784 has been marked as a duplicate of this bug. ***
*** Bug 208674 has been marked as a duplicate of this bug. ***
*** Bug 208758 has been marked as a duplicate of this bug. ***
*** Bug 209028 has been marked as a duplicate of this bug. ***
*** Bug 209126 has been marked as a duplicate of this bug. ***
*** Bug 210476 has been marked as a duplicate of this bug. ***
*** Bug 210578 has been marked as a duplicate of this bug. ***
*** Bug 211204 has been marked as a duplicate of this bug. ***
I have got this error, too: debian/sid (sidux) kmail: 4:4.3.2-1 libqt4: 4.5.3-4 What I was doing: Testing the filtering options in kmail to see, if it is worth to change to kmail from thunderbird. I have a local imap server and filtered some mails (from bugs.kde.org) on manual apply from one imap folder to another. Then I clicked around a little bit in kmail and decided to quit kmail. kmail exited an the Crash Reporting Assistant did show up: Application: KMail (kmail), signal: Segmentation fault [KCrash Handler] #6 0x00000000 in ?? () #7 0xb4b7450e in qDeleteAll<KIO::SchedulerPrivate::ProtocolInfoDict> () at /usr/include/qt4/QtCore/qalgorithms.h:358 #8 ~SchedulerPrivate () at ../../kio/kio/scheduler.cpp:103 #9 destroy () at ../../kio/kio/scheduler.cpp:209 #10 0xb4aa9cfb in ~KCleanUpGlobalStatic (this=0xb4c73214, __in_chrg=<value optimized out>) at ../../kdecore/kernel/kglobal.h:62 #11 0xb5afb589 in exit () from /lib/i686/cmov/libc.so.6 #12 0xb5ae17ad in __libc_start_main () from /lib/i686/cmov/libc.so.6 #13 0x08049c71 in _start ()
*** Bug 211675 has been marked as a duplicate of this bug. ***
*** Bug 212120 has been marked as a duplicate of this bug. ***
I have the same bug, and have found the reason. The reason is as follows: In the middle of the function killAllJobs() (in kmaacctimap.cpp), if the following code were excuted if (mSlave && mapJobData.begin() != mapJobData.end()) { mSlave->kill(); mSlave = 0; } then the last portion of this function if ( disconnectSlave && slave() ) { KIO::Scheduler::disconnectSlave( slave() ); mSlave = 0; } is not executed because of slave() == NULL. This implies that slave() is never disconnected, which causes crashes in the kio scheduler when kmail exists. If you fix it so that slave() is disconnected, then the crash goes away, at least, in my case.
*** Bug 212648 has been marked as a duplicate of this bug. ***
SVN commit 1043972 by tmcguire: Backport r1043961 by tmcguire from trunk to the 4.3 branch: Possibly fix a crash when exiting KMail, because the scheduler was never told to disconnect the slave. Bug figured out by honda@math.sci.hokudai.ac.jp, thanks very much! BUG: 199375 M +9 -7 kmacctimap.cpp WebSVN link: http://websvn.kde.org/?view=rev&revision=1043972
*** Bug 213104 has been marked as a duplicate of this bug. ***
*** Bug 213521 has been marked as a duplicate of this bug. ***
*** Bug 213564 has been marked as a duplicate of this bug. ***
*** Bug 213948 has been marked as a duplicate of this bug. ***
*** Bug 214767 has been marked as a duplicate of this bug. ***
*** Bug 214766 has been marked as a duplicate of this bug. ***
*** Bug 214753 has been marked as a duplicate of this bug. ***
*** Bug 214779 has been marked as a duplicate of this bug. ***
*** Bug 215095 has been marked as a duplicate of this bug. ***
*** Bug 215107 has been marked as a duplicate of this bug. ***
*** Bug 215245 has been marked as a duplicate of this bug. ***
*** Bug 215326 has been marked as a duplicate of this bug. ***
*** Bug 215315 has been marked as a duplicate of this bug. ***
*** Bug 215602 has been marked as a duplicate of this bug. ***
*** Bug 215685 has been marked as a duplicate of this bug. ***
*** Bug 215991 has been marked as a duplicate of this bug. ***
*** Bug 216119 has been marked as a duplicate of this bug. ***
*** Bug 216265 has been marked as a duplicate of this bug. ***
*** Bug 215820 has been marked as a duplicate of this bug. ***
*** Bug 216451 has been marked as a duplicate of this bug. ***
*** Bug 216840 has been marked as a duplicate of this bug. ***
*** Bug 217545 has been marked as a duplicate of this bug. ***
*** Bug 217162 has been marked as a duplicate of this bug. ***
*** Bug 216911 has been marked as a duplicate of this bug. ***
*** Bug 218358 has been marked as a duplicate of this bug. ***
*** Bug 218701 has been marked as a duplicate of this bug. ***
*** Bug 218908 has been marked as a duplicate of this bug. ***
Reopening as it seems the issue is not fully fixed.
*** Bug 220178 has been marked as a duplicate of this bug. ***
For me, a fix in 4.3.4 completely solve the problem. However I doubt if some racing conditions exist. In fact, before the fix, the order is mSlave->kill() -> mSlave = NULL -> disconnectSlave(mSlave), after the fix, disconnectSlave(mSlave) -> mSlave->kill() -> mSlave = NULL. The problem here is that all the other routines in kmail determines their behaviors by checking if mSlave is NULL. So before mSlave becoming null, there is a slightly big period where a race occures. Therefore it might be better to make mSlave NULL first!, I think.
*** Bug 220418 has been marked as a duplicate of this bug. ***
*** Bug 220764 has been marked as a duplicate of this bug. ***
*** Bug 221002 has been marked as a duplicate of this bug. ***
*** Bug 221000 has been marked as a duplicate of this bug. ***
*** Bug 220999 has been marked as a duplicate of this bug. ***
*** Bug 221601 has been marked as a duplicate of this bug. ***
*** Bug 221801 has been marked as a duplicate of this bug. ***
Created attachment 39755 [details] New crash information added by DrKonqi Just like 221000 this happened to me when I unminimized kmail (because it had new mail because I've been reporting a cascade of crashes since upgrading to 4.4rc1) and then the Akonadi thing popped up a complaint about a bunch of trouble it was having. When I closed that kmail crashed before I could click "Inbox". I was not attempting to exit kmail at the time, so the bug title is a bit off, but the backtrace looks about the same to me.
*** Bug 222174 has been marked as a duplicate of this bug. ***
Created attachment 39980 [details] New crash information added by DrKonqi kmail crashed after closing akonadi status report
*** Bug 223373 has been marked as a duplicate of this bug. ***
Created attachment 40090 [details] New crash information added by DrKonqi Same crash on kubuntu 9.10 (64 bit) with kde 4.4 rc1 from kubuntu beta ppa.
*** Bug 223750 has been marked as a duplicate of this bug. ***
Scheduler shutdown (and in fact the whole scheduler) works differently now, so this may or may not be fixed in SVN trunk. Any reports of similar crashes with trunk would be very interesting. If the new scheduler turns out to be stable it might make sense to backport it to 4.4, but it's a big change. Currently there are no open bugs against it.
*** Bug 225196 has been marked as a duplicate of this bug. ***
*** Bug 225138 has been marked as a duplicate of this bug. ***
What should I do about this bug? KMail is crashing when I run it the first time and KAddressBook does not display any addressbook entries. What information do I need to provide? It's not clear from this bug. Please fix this bug. Regards, -- Shlomi Fish
<offtopic> @Shlomi Fish: Your KMail bug is a different one... (this report is about a crash on exit). Please use the report assistant in the Crash Handler dialog to report a new crash. About the Akregator issue: please read http://tokoe-kde.blogspot.com/2009/09/kaddressbook-and-akonadi-in-kde-44.html , http://forum.kde.org/viewtopic.php?f=20&t=85382 and bug 224732 Regards </offtopic>
That's probably not much help, but Ubuntu VLC has a lot of crash reports that look awfully similar to this one. See http://bugs.launchpad.net/bugs/408719 and all its dups. VLC crashes after returning from main(), while trying to delete the KIO scheduler. Unfortunately, it is not clear to me why KIO is there at all (VLC does not use it, but only pure Qt4) and I do not know how to reproduce the problem.
(In reply to comment #81) > That's probably not much help, but Ubuntu VLC has a lot of crash reports that > look awfully similar to this one. See http://bugs.launchpad.net/bugs/408719 and > all its dups. > > VLC crashes after returning from main(), while trying to delete the KIO > scheduler. Unfortunately, it is not clear to me why KIO is there at all (VLC > does not use it, but only pure Qt4) and I do not know how to reproduce the > problem. I bet that this is due to the KDE file open dialog.
(In reply to comment #82) > I bet that this is due to the KDE file open dialog. Right! On Kubuntu 9.10, start VLC, go to Media / Open file, cancel, and exit. 100% segmentation fault with the same backtrace. I don't fully understand the static global pattern that the KIO scheduler appears to use. Is the cleanup always done at exit? or could the crash imply someone is "leaking" a reference to KIO, which only then falls back to an exit cleanup handler?
*** Bug 231865 has been marked as a duplicate of this bug. ***
*** Bug 232190 has been marked as a duplicate of this bug. ***
*** Bug 230378 has been marked as a duplicate of this bug. ***
*** Bug 229166 has been marked as a duplicate of this bug. ***
*** Bug 223881 has been marked as a duplicate of this bug. ***
*** Bug 225860 has been marked as a duplicate of this bug. ***
*** Bug 223803 has been marked as a duplicate of this bug. ***
*** Bug 225599 has been marked as a duplicate of this bug. ***
*** Bug 228249 has been marked as a duplicate of this bug. ***
*** Bug 232368 has been marked as a duplicate of this bug. ***
*** Bug 222411 has been marked as a duplicate of this bug. ***
*** Bug 222265 has been marked as a duplicate of this bug. ***
*** Bug 234206 has been marked as a duplicate of this bug. ***
*** Bug 236038 has been marked as a duplicate of this bug. ***
*** Bug 235662 has been marked as a duplicate of this bug. ***
*** Bug 235655 has been marked as a duplicate of this bug. ***
*** Bug 205083 has been marked as a duplicate of this bug. ***
If no one see this bug after the scheduler rewrite in KDE 4.5. This bug should be closed as fixed.
*** Bug 209374 has been marked as a duplicate of this bug. ***
Closing based on comment #101