Bug 145303 - [crash] re-filtering of mailing lists, duplication of filtered messages
Summary: [crash] re-filtering of mailing lists, duplication of filtered messages
Status: RESOLVED NOT A BUG
Alias: None
Product: kmail
Classification: Applications
Component: filtering (show other bugs)
Version: 1.9.6
Platform: unspecified Linux
: NOR crash
Target Milestone: ---
Assignee: kdepim bugs
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2007-05-11 08:43 UTC by Daniel Franke
Modified: 2007-05-12 14:48 UTC (History)
0 users

See Also:
Latest Commit:
Version Fixed In:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Daniel Franke 2007-05-11 08:43:52 UTC
Version:           1.9.6 (using KDE 3.5.6, Gentoo)
Compiler:          Target: i686-pc-linux-gnu
OS:                Linux (i686) release 2.6.17.13

Two high-volume mailing-lists (A and B), both filtered into their respective directory. Messages, that are sent to both, shall be filtered into the directory of ML A. Since the filter for B was set up after that of A, it is applied first, thus mails sent two both lists go to directory B. Change the ordering of filters so that A comes first (move to top of filter list). Select all mails in directory B (6000+ in my case), apply filter. Filter is applied to most mails, then kmail chrashes. The mails in folder B are duplicated (now 12000+). Folder A grew by a couple of hundred, hence, no mail seems to be lost.

Unfortunately, I didn't keep the backtrace. I also hesitate to attempt to reproduce this ... 

Any suggestions how to get rid of 6000 duplicated mails?
Comment 1 Thomas McGuire 2007-05-12 14:05:07 UTC
I guess this issue has already been reported, but without backtrace I can't be really sure. Therefore I'll close this bug report as INVALID. If you have a backtrace which is different from those in bug 135376, feel free to reopen this report.

Crash after filtering: bug 135376
Duplicate messages after filtering: bug 95064
Comment 2 Daniel Franke 2007-05-12 14:34:08 UTC
Regarding #95064:
My problem was not related to IMAP. The mails were fetched via POP3 and stored locally - then refiltered by applying the filter command manually to mails in a local folder.

Nevertheless, it looks like a dupe of #135376.
Comment 3 Daniel Franke 2007-05-12 14:37:13 UTC
Couldn't stop myself. Here's a backtrace (seems to be the same as shown in #135376, comment 5):

Using host libthread_db library "/lib/libthread_db.so.1".
[Thread debugging using libthread_db enabled]
[New Thread -1246140096 (LWP 6489)]
[New Thread -1283781744 (LWP 6493)]
[New Thread -1275389040 (LWP 6492)]
[New Thread -1266996336 (LWP 6491)]
[New Thread -1258603632 (LWP 6490)]
[KCrash handler]
#6  0xb532eca0 in KMFolder::find () from /usr/kde/3.5/lib/libkmailprivate.so
#7  0xb52f88a5 in KMHeaders::applyFiltersOnMsg ()
   from /usr/kde/3.5/lib/libkmailprivate.so
#8  0xb548106e in KMMainWidget::slotApplyFilters ()
   from /usr/kde/3.5/lib/libkmailprivate.so
#9  0xb549a7be in KMMainWidget::qt_invoke ()
   from /usr/kde/3.5/lib/libkmailprivate.so
#10 0xb64caa47 in QObject::activate_signal (this=0x85541a0, clist=0x85543f0, 
    o=0xbfa0690c) at kernel/qobject.cpp:2356
#11 0xb64cb7e0 in QObject::activate_signal (this=0x85541a0, signal=2)
    at kernel/qobject.cpp:2325
#12 0xb6fb49ae in KAction::activated (this=0x85541a0) at kaction.moc:176
#13 0xb6fb5138 in KAction::slotActivated (this=0x85541a0) at kaction.cpp:1102
#14 0xb6fb56aa in KAction::qt_invoke (this=0x85541a0, _id=15, _o=0xbfa06a3c)
    at kaction.moc:218
#15 0xb64caa47 in QObject::activate_signal (this=0x8a1b688, clist=0x82933d8, 
    o=0xbfa06a3c) at kernel/qobject.cpp:2356
#16 0xb64cb7e0 in QObject::activate_signal (this=0x8a1b688, signal=2)
    at kernel/qobject.cpp:2325
#17 0xb6d4b59e in KAccelPrivate::menuItemActivated (this=0x8a1b688)
    at kaccelprivate.moc:110
#18 0xb6d4c9a4 in KAccelPrivate::emitActivatedSignal (this=0x8a1b688, 
    pAction=0x8a387c0) at kaccel.cpp:400
#19 0xb6d4d011 in KAccelPrivate::eventFilter (this=0x8a1b688, 
    pEvent=0xbfa06f7c) at kaccel.cpp:370
#20 0xb64c9e9f in QObject::activate_filters (this=0x80d4608, e=0xbfa06f7c)
    at kernel/qobject.cpp:903
#21 0xb64c9f40 in QObject::event (this=0x80d4608, e=0xbfa06f7c)
    at kernel/qobject.cpp:735
#22 0xb650b626 in QWidget::event (this=0x80d4608, e=0xbfa06f7c)
    at kernel/qwidget.cpp:4659
#23 0xb65f7cbe in QMainWindow::event (this=0x80d4608, e=0xbfa06f7c)
    at widgets/qmainwindow.cpp:1687
#24 0xb64561d3 in QApplication::internalNotify (this=0xbfa074c8, 
    receiver=0x80d4608, e=0xbfa06f7c) at kernel/qapplication.cpp:2635
#25 0xb64582e2 in QApplication::notify (this=0xbfa074c8, receiver=0x84cba10, 
    e=0xbfa06f7c) at kernel/qapplication.cpp:2392
#26 0xb6cdd13d in KApplication::notify (this=0xbfa074c8, receiver=0x84cba10, 
    event=0xbfa06f7c) at kapplication.cpp:550
#27 0xb789faba in QApplication::sendEvent (receiver=0x84cba10, 
    event=0xbfa06f7c) at /usr/qt/3/include/qapplication.h:496
#28 0xb6d4dc37 in KAccelEventHandler::x11Event (this=0x80f6bd8, 
    pEvent=0xbfa07328) at kaccel.cpp:144
#29 0xb6cde10f in KAppX11HackWidget::publicx11Event (this=0x80f6bd8, 
    e=0xbfa07328) at kapplication.cpp:1659
#30 0xb6cd4d2b in KApplication::x11EventFilter (this=0xbfa074c8, 
    _event=0xbfa07328) at kapplication.cpp:1729
#31 0xb63c9f3b in qt_x11EventFilter (ev=0xbfa07328)
    at kernel/qapplication_x11.cpp:387
#32 0xb63dc503 in QApplication::x11ProcessEvent (this=0xbfa074c8, 
    event=0xbfa07328) at kernel/qapplication_x11.cpp:3315
#33 0xb63f5926 in QEventLoop::processEvents (this=0x809c5b8, flags=4)
    at kernel/qeventloop_x11.cpp:192
#34 0xb64741ed in QEventLoop::enterLoop (this=0x809c5b8)
    at kernel/qeventloop.cpp:198
#35 0xb647400e in QEventLoop::exec (this=0x809c5b8)
    at kernel/qeventloop.cpp:145
#36 0xb6457e97 in QApplication::exec (this=0xbfa074c8)
    at kernel/qapplication.cpp:2758
#37 0x08058e8c in main ()
Comment 4 Thomas McGuire 2007-05-12 14:48:24 UTC
Yes, the backtrace is the same.
This probably also has to do something with the duplication of messages, probably fixed when the crash gets fixed.