| Summary: | KMail crashes if nepomuk is enabled | ||
|---|---|---|---|
| Product: | [Applications] kmail2 | Reporter: | Unknown <null> |
| Component: | general | Assignee: | kdepim bugs <kdepim-bugs> |
| Status: | RESOLVED FIXED | ||
| Severity: | crash | CC: | anti.desitter, didier.richard, faure, franz.trischberger, johu, k.woelfel, kde-bugs, max.lin, montel, mschiff, rdieter, robert, siocnarf, sven.burmeister, than, zerg |
| Priority: | NOR | ||
| Version: | 4.8.4 | ||
| Target Milestone: | --- | ||
| Platform: | openSUSE | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed In: | 4.8.5 | |
| Sentry Crash Report: | |||
| Bug Depends on: | 301453 | ||
| Bug Blocks: | |||
| Attachments: |
New crash information added by DrKonqi
New crash information added by DrKonqi New crash information added by DrKonqi New crash information added by DrKonqi |
||
|
Description
Unknown
2011-07-13 15:56:50 UTC
*** Bug 278944 has been marked as a duplicate of this bug. *** *** Bug 289128 has been marked as a duplicate of this bug. *** *** Bug 291293 has been marked as a duplicate of this bug. *** *** Bug 299355 has been marked as a duplicate of this bug. *** Don't crash if nepomuk switched off Confirmed. Disabling nepomuk in systemsettings and killing all nepomuk* processes make me able to use kmail again (current 4.8 branch). Thanks for the hint! ?????? do you see this bactrace ???? do you see nepomuk here ???? Please report on good bug report, and not just read "it's crash" ok it's same for me it's crash. So no nepomuk is not related to this crash. No nepomuk in backtrace. Backtrace same as in description. When turn nepomuk on #299357 appear too. (In reply to comment #7) > ?????? do you see this bactrace ???? do you see nepomuk here ???? > > Please report on good bug report, and not just read "it's crash" ok it's > same for me it's crash. > > So no nepomuk is not related to this crash. Hi Laurent, Its true that there is nothing about nepomuk in the BT. But it is also true, that this crash only occurs if nepomuk is enabled. BTW: Enabling nepomuk also make akonadi nepomuk feeder and akonadi mail agent crash immediately. Are you interested in those BTs as well? Thanks -Marc (In reply to comment #10) > (In reply to comment #7) > > ?????? do you see this bactrace ???? do you see nepomuk here ???? > > > > Please report on good bug report, and not just read "it's crash" ok it's > > same for me it's crash. > > > > So no nepomuk is not related to this crash. > > Hi Laurent, > > Its true that there is nothing about nepomuk in the BT. But it is also true, > that this crash only occurs if nepomuk is enabled. > BTW: Enabling nepomuk also make akonadi nepomuk feeder and akonadi mail > agent crash immediately. Are you interested in those > BTs as well? > > Thanks > -Marc Yes please put bt Thanks (In reply to comment #11) > please put bt http://bugs.kde.org/299357 *** Bug 300203 has been marked as a duplicate of this bug. *** > Yes please put bt > Thanks I only can reproduce the nepomuk feeder crash atm. (Bug 299357) Steps to reproduce: 1. akonadictl stop (wait until 'akonadictl status' tells that its stopped) 2. enable nepomuk 3. akonadictl start -> crash No problem after building soprano and strigi with clucene-core-2.3.3.4 (was clucene-0.9.21) But only with mail indexing. When turn file indexing on, problem occur again. Reverting kdelibs/nepomuk/* code to v4.8.3 tag fixes the problem Created attachment 71664 [details]
New crash information added by DrKonqi
kmail (4.8.4) on KDE Platform 4.8.4 (4.8.4) using Qt 4.8.2
- What I was doing when the application crashed:
attempt to refresh my gmail imap inbox using kmail from kdepim-4.8.4
-- Backtrace (Reduced):
#7 ~KSharedPtr (this=0xcb7050, __in_chrg=<optimized out>) at /usr/src/debug/kdelibs-4.8.4/kdecore/util/ksharedptr.h:90
#8 ~ParsingData (this=0xcb7050, __in_chrg=<optimized out>) at /usr/src/debug/kdelibs-4.8.4/kdecore/services/ktraderparse.cpp:40
#9 qThreadStorage_deleteData<KTraderParse::ParsingData> (d=0xcb7050) at /usr/include/QtCore/qthreadstorage.h:97
#10 QThreadStorage<KTraderParse::ParsingData*>::deleteData (x=0xcb7050) at /usr/include/QtCore/qthreadstorage.h:140
#11 0x00000039874793e0 in QThreadStorageData::set (this=<optimized out>, p=0x19bef40) at thread/qthreadstorage.cpp:165
confirmed that disabling nepomuk completely and crash no longer occurs. Seems to still happen even if I disable either (or both) of Nepomuk File Indexer, Email Indexer updating summary to match findings of last 2 folks (including me) Don't crash with lastest 4.8 branch, but akonadi_nepomuk_feeder still "System busy, indexing suspended" Created attachment 71675 [details]
New crash information added by DrKonqi
kmail (4.8.4) on KDE Platform 4.8.4 (4.8.4) using Qt 4.8.2
- What I was doing when the application crashed:
crash by selecting an incoming email after boot
- Custom settings of the application:
2 imap account and 2 pop
4 smtp account
default configuration
-- Backtrace (Reduced):
#6 0x00007feedfeda470 in QThreadStorageData::set(void*) () from /usr/lib/libQtCore.so.4
[...]
#8 0x00007feee17846f2 in KServiceTypeTrader::applyConstraints(QList<KSharedPtr<KService> >&, QString const&) () from /usr/lib/libkdecore.so.5
#9 0x00007feee178487c in KServiceTypeTrader::defaultOffers(QString const&, QString const&) const () from /usr/lib/libkdecore.so.5
#10 0x00007feee1784c31 in KServiceTypeTrader::query(QString const&, QString const&) const () from /usr/lib/libkdecore.so.5
[...]
#12 0x00007feed1920a3a in KEmoticons::KEmoticons() () from /usr/lib/libkemoticons.so.4
have try without nepomuk, but crash again I can confirm that turning off nepomuk stops kmail from crashing. kontact does not crash when nepomuk is turned on and kmail gets loaded, therefor it does not stop correctly - it keeps running and prevents starting a new instance. Here is the backtrace of the locked up instance: (gdb) bt #0 syscall () at ../sysdeps/unix/sysv/linux/x86_64/syscall.S:39 #1 0x00007f507b24771b in _q_futex (val2=0, addr2=0x0, timeout=0x0, val=2, op=0, addr=0x27fe930) at thread/qmutex_unix.cpp:99 #2 QMutexPrivate::wait (this=0x27fe930, timeout=<optimized out>) at thread/qmutex_unix.cpp:113 #3 0x00007f507b2434fd in QMutex::lockInternal (this=<optimized out>) at thread/qmutex.cpp:450 #4 0x00007f506e339c5a in lockInline (this=0x27fe5e8) at /usr/include/qt4/QtCore/qmutex.h:198 #5 QMutexLocker (m=0x27fe5e8, this=<synthetic pointer>) at /usr/include/qt4/QtCore/qmutex.h:109 #6 Soprano::Client::SocketHandler::~SocketHandler (this=0x2800a90, __in_chrg=<optimized out>) at /var/tmp/paludis/dev-libs-soprano-2.7.6/work/soprano-2.7.6/client/clientconnection.cpp:58 #7 0x00007f506e339d29 in Soprano::Client::SocketHandler::~SocketHandler (this=0x2800a90, __in_chrg=<optimized out>) at /var/tmp/paludis/dev-libs-soprano-2.7.6/work/soprano-2.7.6/client/clientconnection.cpp:61 #8 0x00007f507b246c0c in QThreadStorageData::set (this=0x2436230, p=0x27ff7e0) at thread/qthreadstorage.cpp:165 #9 0x00007f506e33774f in qThreadStorage_setLocalData<Soprano::Client::SocketHandler> (d=..., t=<optimized out>) at /usr/include/qt4/QtCore/qthreadstorage.h:92 #10 setLocalData (t=0x27ff7e0, this=<optimized out>) at /usr/include/qt4/QtCore/qthreadstorage.h:155 #11 Soprano::Client::ClientConnection::socketForCurrentThread (this=0x2511020) at /var/tmp/paludis/dev-libs-soprano-2.7.6/work/soprano-2.7.6/client/clientconnection.cpp:95 #12 0x00007f506e3377b9 in Soprano::Client::ClientConnection::connectInCurrentThread (this=<optimized out>) at /var/tmp/paludis/dev-libs-soprano-2.7.6/work/soprano-2.7.6/client/clientconnection.cpp:800 #13 0x00007f506e336d8a in Soprano::Client::LocalSocketClient::connect (this=0x27fda78, name="/tmp/ksocket-franz/nepomuk-socket") at /var/tmp/paludis/dev-libs-soprano-2.7.6/work/soprano-2.7.6/client/localsocketclient.cpp:141 #14 0x00007f50728a1157 in init (forced=true, this=0x27fda50) at /var/tmp/paludis/kde-base-kdelibs-4.8.4/work/kdelibs-4.8.4/nepomuk/core/nepomukmainmodel.cpp:103 #15 Nepomuk::MainModel::init (this=0x27fd380) at /var/tmp/paludis/kde-base-kdelibs-4.8.4/work/kdelibs-4.8.4/nepomuk/core/nepomukmainmodel.cpp:177 #16 0x00007f5072899f20 in Nepomuk::ResourceManager::init (this=0x27fcc20) at /var/tmp/paludis/kde-base-kdelibs-4.8.4/work/kdelibs-4.8.4/nepomuk/core/resourcemanager.cpp:323 #17 0x00007f507289a105 in Nepomuk::ResourceManager::ResourceManager (this=0x27fcc20) at /var/tmp/paludis/kde-base-kdelibs-4.8.4/work/kdelibs-4.8.4/nepomuk/core/resourcemanager.cpp:286 #18 0x00007f507289a155 in Nepomuk::ResourceManager::instance () at /var/tmp/paludis/kde-base-kdelibs-4.8.4/work/kdelibs-4.8.4/nepomuk/core/resourcemanager.cpp:311 #19 0x00007f50728b0283 in Nepomuk::ResourceWatcher::stop (this=0x28061b0) at /var/tmp/paludis/kde-base-kdelibs-4.8.4/work/kdelibs-4.8.4/nepomuk/core/resourcewatcher.cpp:140 #20 0x00007f50728b02e0 in Nepomuk::ResourceWatcher::~ResourceWatcher (this=0x28061b0, __in_chrg=<optimized out>) at /var/tmp/paludis/kde-base-kdelibs-4.8.4/work/kdelibs-4.8.4/nepomuk/core/resourcewatcher.cpp:79 #21 0x00007f50728b0339 in Nepomuk::ResourceWatcher::~ResourceWatcher (this=0x28061b0, __in_chrg=<optimized out>) at /var/tmp/paludis/kde-base-kdelibs-4.8.4/work/kdelibs-4.8.4/nepomuk/core/resourcewatcher.cpp:81 #22 0x00007f507b3571e1 in QObjectPrivate::deleteChildren (this=0x27fc8f0) at kernel/qobject.cpp:1908 #23 0x00007f507b35d20c in QObject::~QObject (this=0x27f9530, __in_chrg=<optimized out>) at kernel/qobject.cpp:927 #24 0x00007f507289d4f9 in Nepomuk::ResourceManager::~ResourceManager (this=0x27f9530, __in_chrg=<optimized out>) at /var/tmp/paludis/kde-base-kdelibs-4.8.4/work/kdelibs-4.8.4/nepomuk/core/resourcemanager.cpp:299 #25 0x00007f507b3571e1 in QObjectPrivate::deleteChildren (this=0x1f98da0) at kernel/qobject.cpp:1908 #26 0x00007f507b35d20c in QObject::~QObject (this=0x7fff16024130, __in_chrg=<optimized out>) at kernel/qobject.cpp:927 #27 0x00007f507a6fa7f7 in QApplication::~QApplication (this=0x7fff16024130, __in_chrg=<optimized out>) at kernel/qapplication.cpp:1095 #28 0x0000000000403bff in main (argc=1, argv=0x7fff16024758) at /var/tmp/paludis/kde-base-kontact-4.8.4/work/kontact-4.8.4/kontact/src/main.cpp:221 It hangs in a locked mutex. nepomuk was not correctly stopped (my mistake), kmail works effectively if we deactivate it thank you kmail is not the only affected application: https://bugs.kde.org/show_bug.cgi?id=301453 Solution: Update to soprano-2.7.56 (current beta1 :/) *** Bug 301602 has been marked as a duplicate of this bug. *** Created attachment 71757 [details]
New crash information added by DrKonqi
kmail (4.8.4) on KDE Platform 4.8.4 (4.8.4) "release 511" using Qt 4.8.2
- What I was doing when the application crashed:
I opened kmail and selected one email for reading
- Custom settings of the application:
Other than the original bug description, it also happens for me if, I don't set the message pane next to the message list. It happens as soon as I doubleclick a message for reading.
-- Backtrace (Reduced):
#7 ~KSharedPtr (this=0xf85130, __in_chrg=<optimized out>) at /usr/src/debug/kdelibs-4.8.4/kdecore/util/ksharedptr.h:90
#8 ~ParsingData (this=0xf85130, __in_chrg=<optimized out>) at /usr/src/debug/kdelibs-4.8.4/kdecore/services/ktraderparse.cpp:40
#9 qThreadStorage_deleteData<KTraderParse::ParsingData> (d=0xf85130) at /usr/include/QtCore/qthreadstorage.h:97
#10 QThreadStorage<KTraderParse::ParsingData*>::deleteData (x=0xf85130) at /usr/include/QtCore/qthreadstorage.h:140
#11 0x00007f152b6115ed in QThreadStorageData::set (this=0x1557690, p=0x1a645d0) at thread/qthreadstorage.cpp:165
Created attachment 71758 [details]
New crash information added by DrKonqi
kmail (4.8.4) on KDE Platform 4.8.4 (4.8.4) "release 511" using Qt 4.8.2
Finally, my here prvious reported crash seems not to be related to nepomuk, because I deactivated nepomuk and the same crash happend again.
-- Backtrace (Reduced):
#6 fullLookupForWriting<char const*, WTF::HashSetTranslatorAdapter<WTF::StringImpl*, WTF::HashTraits<WTF::StringImpl*>, char const*, WTF::CStringTranslator> > (key=<synthetic pointer>, this=0x7f67b89b9e10) at ../../../Source/JavaScriptCore/wtf/text/AtomicString.cpp:368
#7 addPassingHashCode<char const*, char const*, WTF::HashSetTranslatorAdapter<WTF::StringImpl*, WTF::HashTraits<WTF::StringImpl*>, char const*, WTF::CStringTranslator> > (this=0x7f67b89b9e10, extra=<synthetic pointer>, key=<synthetic pointer>) at ../../../Source/JavaScriptCore/wtf/HashTable.h:726
#8 add<char const*, WTF::CStringTranslator> (this=<optimized out>, value=<synthetic pointer>) at ../../../Source/JavaScriptCore/wtf/HashSet.h:189
#9 addToStringTable<char const*, WTF::CStringTranslator> (value=<synthetic pointer>) at ../../../Source/JavaScriptCore/wtf/text/AtomicString.cpp:80
#10 WTF::AtomicString::add (c=0x7f67bb89a020 "#text") at ../../../Source/JavaScriptCore/wtf/text/AtomicString.cpp:130
My theory would be that nepomuk corrups memory in QThreadStorage stuff, which then makes KTraderParse crash (it uses its own QThreadStorage, but maybe Qt puts these close to each other in memory). A bit far-fetched though, only a valgrind log would tell. In other news, there have been fixes in nepomuk very recently, to fix crashes with older versions of soprano, so maybe it fixes this. http://commits.kde.org/kdelibs/bde5aad1e9f7bfb6f8d5c60197797de4a315158d http://commits.kde.org/kdelibs/76e0376d7582cfe5b898c56bf3779ca2279dace8 (In reply to comment #29) Sorry, my previous comment was wrong. After disabling nepomuk, I did not kill remaining nepomuk related processes, that was why the crash still happened. When I disable nepomuk and reboot, than the crash does definitely not happen anymore. I would also like to not that the crash happened after updating from 4.8.3 to 4.8.4 on OpenSuSe. > Created attachment 71758 [details] > New crash information added by DrKonqi > > kmail (4.8.4) on KDE Platform 4.8.4 (4.8.4) "release 511" using Qt 4.8.2 > > Finally, my here prvious reported crash seems not to be related to nepomuk, > because I deactivated nepomuk and the same crash happend again. > > -- Backtrace (Reduced): > #6 fullLookupForWriting<char const*, > WTF::HashSetTranslatorAdapter<WTF::StringImpl*, > WTF::HashTraits<WTF::StringImpl*>, char const*, WTF::CStringTranslator> > > (key=<synthetic pointer>, this=0x7f67b89b9e10) at > ../../../Source/JavaScriptCore/wtf/text/AtomicString.cpp:368 > #7 addPassingHashCode<char const*, char const*, > WTF::HashSetTranslatorAdapter<WTF::StringImpl*, > WTF::HashTraits<WTF::StringImpl*>, char const*, WTF::CStringTranslator> > > (this=0x7f67b89b9e10, extra=<synthetic pointer>, key=<synthetic pointer>) at > ../../../Source/JavaScriptCore/wtf/HashTable.h:726 > #8 add<char const*, WTF::CStringTranslator> (this=<optimized out>, > value=<synthetic pointer>) at > ../../../Source/JavaScriptCore/wtf/HashSet.h:189 > #9 addToStringTable<char const*, WTF::CStringTranslator> (value=<synthetic > pointer>) at ../../../Source/JavaScriptCore/wtf/text/AtomicString.cpp:80 > #10 WTF::AtomicString::add (c=0x7f67bb89a020 "#text") at > ../../../Source/JavaScriptCore/wtf/text/AtomicString.cpp:130 See https://bugs.kde.org/show_bug.cgi?id=301453 for more information. Deactivating Nepomuk also "solved" my KMail-Crashing problems. Closing based on comment #32 |