Bug 112128

*** Bug 115361 has been marked as a duplicate of this bug. ***

I came here from bug #115361.

The crash below happened while moving about 5 selected messages from my Inbox to a subfolder of the same IMAP account (below Inbox).

(no debugging symbols found)
Using host libthread_db library "/lib/tls/i686/cmov/libthread_db.so.1".
(no debugging symbols found)
`system-supplied DSO at 0xffffe000' has disappeared; keeping its symbols.
(no debugging symbols found)
[Thread debugging using libthread_db enabled]
[New Thread -1248106816 (LWP 12206)]
[KCrash handler]
#4  0x00000048 in ?? ()
#5  0xb7c53781 in KMHeaders::msgRemoved () from /usr/lib/libkmailprivate.so
#6  0xb7e78946 in KMHeaders::qt_invoke () from /usr/lib/libkmailprivate.so
#7  0xb61b3929 in QObject::activate_signal () from /usr/lib/libqt-mt.so.3
#8  0xb7c0b009 in KMFolder::msgRemoved () from /usr/lib/libkmailprivate.so
#9  0xb7c5a30d in KMFolder::qt_emit () from /usr/lib/libkmailprivate.so
#10 0xb61b38fd in QObject::activate_signal () from /usr/lib/libqt-mt.so.3
#11 0xb7c0beb9 in FolderStorage::msgRemoved ()
   from /usr/lib/libkmailprivate.so
#12 0xb7e5c418 in FolderStorage::take () from /usr/lib/libkmailprivate.so
#13 0xb7e5c696 in KMFolderImap::take () from /usr/lib/libkmailprivate.so
#14 0xb7bfc890 in FolderStorage::take () from /usr/lib/libkmailprivate.so
#15 0xb7da0fa3 in KMFolderImap::take () from /usr/lib/libkmailprivate.so
#16 0xb7c2171c in KMFolder::take () from /usr/lib/libkmailprivate.so
#17 0xb7d9e438 in KMFolderImap::addMsgQuiet ()
   from /usr/lib/libkmailprivate.so
#18 0xb7e5fbd5 in KMFolderImap::qt_invoke () from /usr/lib/libkmailprivate.so
#19 0xb61b3929 in QObject::activate_signal () from /usr/lib/libqt-mt.so.3
#20 0xb7c10e3d in KMail::FolderJob::messageCopied ()
   from /usr/lib/libkmailprivate.so
#21 0xb7da7d19 in KMail::ImapJob::slotCopyMessageResult ()
   from /usr/lib/libkmailprivate.so
#22 0xb7dad603 in KMail::ImapJob::qt_invoke ()
   from /usr/lib/libkmailprivate.so
#23 0xb61b3929 in QObject::activate_signal () from /usr/lib/libqt-mt.so.3
#24 0xb6d459fa in KIO::Job::result (this=0x914aca8, t0=0x914aca8)
    at jobclasses.moc:162
#25 0xb6d966da in KIO::Job::emitResult (this=0x914aca8) at job.cpp:222
#26 0xb6d9684b in KIO::SimpleJob::slotFinished (this=0x914aca8) at job.cpp:570
#27 0xb6d983e5 in KIO::SimpleJob::qt_invoke (this=0x914aca8, _id=7, 
    _o=0xbfc84eac) at jobclasses.moc:417
#28 0xb61b3929 in QObject::activate_signal () from /usr/lib/libqt-mt.so.3
#29 0xb61b43c4 in QObject::activate_signal () from /usr/lib/libqt-mt.so.3
#30 0xb6d40d59 in KIO::SlaveInterface::finished (this=0x85755a0)
    at slaveinterface.moc:226
#31 0xb6daf644 in KIO::SlaveInterface::dispatch (this=0x85755a0, _cmd=104, 
    rawdata=@0xbfc85070) at slaveinterface.cpp:243
#32 0xb6d57720 in KIO::SlaveInterface::dispatch (this=0x85755a0)
    at slaveinterface.cpp:173
#33 0xb6d5c0f4 in KIO::Slave::gotInput (this=0x85755a0) at slave.cpp:300
#34 0xb6d5c289 in KIO::Slave::qt_invoke (this=0x85755a0, _id=4, _o=0xbfc851a8)
    at slave.moc:113
#35 0xb61b3929 in QObject::activate_signal () from /usr/lib/libqt-mt.so.3
#36 0xb61b4238 in QObject::activate_signal () from /usr/lib/libqt-mt.so.3
#37 0xb6513c67 in QSocketNotifier::activated () from /usr/lib/libqt-mt.so.3
#38 0xb61d3faa in QSocketNotifier::event () from /usr/lib/libqt-mt.so.3
#39 0xb614af80 in QApplication::internalNotify () from /usr/lib/libqt-mt.so.3
#40 0xb614b172 in QApplication::notify () from /usr/lib/libqt-mt.so.3
#41 0xb6895eec in KApplication::notify (this=0xbfc85778, receiver=0x8574888, 
    event=0xbfc854f0) at kapplication.cpp:550
#42 0xb60dbdb7 in QApplication::sendEvent () from /usr/lib/libqt-mt.so.3
#43 0xb613cacb in QEventLoop::activateSocketNotifiers ()
   from /usr/lib/libqt-mt.so.3
#44 0xb60efa60 in QEventLoop::processEvents () from /usr/lib/libqt-mt.so.3
#45 0xb6162cfb in QEventLoop::enterLoop () from /usr/lib/libqt-mt.so.3
#46 0xb6162c1e in QEventLoop::exec () from /usr/lib/libqt-mt.so.3
#47 0xb6149c13 in QApplication::exec () from /usr/lib/libqt-mt.so.3
#48 0x0804a3bf in ?? ()
#49 0xbfc85778 in ?? ()
#50 0xbfc85964 in ?? ()
#51 0xbfc85978 in ?? ()
#52 0x0804a36e in ?? ()
#53 0xb7f6fc64 in ?? ()
#54 0xbfc85818 in ?? ()
#55 0xb7f787c9 in _dl_rtld_di_serinfo () from /lib/ld-linux.so.2
#56 0xb59faea2 in __libc_start_main () from /lib/tls/i686/cmov/libc.so.6
#57 0x0804a151 in ?? ()

*** Bug 117581 has been marked as a duplicate of this bug. ***

*** Bug 117536 has been marked as a duplicate of this bug. ***

*** Bug 120807 has been marked as a duplicate of this bug. ***

*** Bug 121510 has been marked as a duplicate of this bug. ***

*** Bug 126104 has been marked as a duplicate of this bug. ***

*** Bug 126999 has been marked as a duplicate of this bug. ***

*** Bug 131313 has been marked as a duplicate of this bug. ***

Looking at all bugs that are linked to this report I'm not sure it's about the same issue.

Ever since upgrading to KDE 3.5 (or even a pre-release SVN snapshot) I've had the problem with online IMAP folders becoming 'invalid' after a while -- the listview became empty, scrolling it caused garbage on the screen and it was no longer possible to select mails. The solution was to switch to another folder and back.

Currently (KDE 3.5.4) this exact same problem seems to still be there, but now causes crashes rather than garbage. Haven't tried 3.5.3, on 3.5.2 it was still the garbage (but no crash).

I have no backtrace (and I'm using SUSE rpms, not a debug build, so there's little to backtrace anyway) but some observations I've made over the past months:

* The problem seems to be a race condition with multiple folder manipulations happening at the same time. Having a folder open passively in the background does not (or hardly?) trigger the problem.

* Slower links tend to amplify the problem, making my race condition hypothesis bigger. At home I have a fast wifi link to my server and hardly hit the issue, from work I am connecting to my server over the relatively slow DSL and hit it frequently.

* A scenario that almost invariably seems to trigger the issue from work (but due to the faster link is hard to reproduce from home): read some mail (using the arrow keys), so mails get marked as 'read'. If at the same time the new mail check happens and new mail has arrived you can be pretty sure that subsequently trying to switch mails or delete them causes the crash (or garbage on older versions). The mail check itself seems safe, it's new mail, in the ACTIVE folder that appears to trigger it, and only if you are marking mails as read/replied/...

* I have no client-side mail filters, all filtering happens on the server through Sieve. This also means that several folders can receive new mail, making the problem likelier to occur.

* FWIW, Used software: SUSE RPMs for 9.2, 9.3 and 10.0 on different systems; KDE versions 3.5.1, 3.5.2 and 3.5.4. Cyrus IMAP on SUSE 9.2 as mail server. No client-side spam- or virus filter. Only one system (home) is gpg-enabled, yet all show the problem. All IMAP connections are online IMAP and use TLS.

For more info you know where to find me ;) (I'm not in a good position to test though, so I hope the above observations help Allen or someone else who knows the flow of operations enough to be helpful.)

*** Bug 142538 has been marked as a duplicate of this bug. ***

This looks like a lot like the crash caused by #126715 - so if whoever had problems with kmail IMAP could try the current 3.5 branch this would be highly appreciated, so that we can fix remaning issues before 3.5.7

*** This bug has been marked as a duplicate of 126715 ***