Bug 334387

Summary: konqueror stops responding, crashes after a while on youtube
Product: [Unmaintained] kdelibs Reporter: rembrand
Component: kdewebkitAssignee: webkit-devel
Status: RESOLVED WORKSFORME    
Severity: crash CC: adawit
Priority: NOR    
Version: 4.12.4   
Target Milestone: ---   
Platform: Fedora RPMs   
OS: Linux   
URL: http://www.youtube.com
Latest Commit: Version Fixed In:
Sentry Crash Report:

Description rembrand 2014-05-05 18:07:20 UTC
Application: konqueror (4.12.4)
KDE Platform Version: 4.12.4
Qt Version: 4.8.6
Operating System: Linux 3.14.2-200.fc20.i686 i686
Distribution (Platform): Fedora RPMs

-- Information about the crash:
- What I was doing when the application crashed:
Browse to the youtube site and start scrolling immediately
using the  scroll bars or start  typing to search for a video.
Do this a bit faster then usual and keep that doing for about 10 seconds.


Reproducible: Always

Steps to Reproduce:
1. 
Browse to the youtube site and start scrolling immediately
using the  scroll bars or start  typing to search for a video.
Do this a bit faster then usual and keep that doing for about 10 seconds.

2.
Scrolling and/or typing becomes slower and slower.
Konqueror stops responding and finally crashes or hangs
permanently. Behavior is reproducible each time. 

3.
When doing things very slow and easy, it works 'better', for 
a longer period of time. Videos can be played and searched for but
after a while konqueror/webkit stops responding too.


Actual Results:  
Konqueror stops responding, hangs permanently and  crashes.
After that konqueror must be started again, no other issues.

Expected Results:  
Without hangs and crashes, browsing youtube sitem searching and watching videos.

- Custom settings of the application:
/usr/lib/mozilla/plugins/libflashplayer.so (Adobe, version 11.2.202.356)
which is a symbolic link to  /usr/lib/flash-plugin/libflashplayer.so
This is the ONLY custom setting I did change.

Notes:
Using the QtWebkit from PyQt4-4.10.2-5.fc20.i686  produces the same result
(!!when used in a PyQt4 program!!).
Using konqueror in webkit mode only.
System is installed from scratch just 2 days ago.

-Versions installed:
qtwebkit-2.3.3-7.fc20.i686
qt-4.8.6-2.fc20.i686
PyQt4-4.10.2-5.fc20.i686
Fedora 20 Linux 3.14.2-200.fc20.i686

This official Fedora KDE spin has been updated with the latest updates 
as of 5 may 2014.
The crash can be reproduced every time.
Comment 1 rembrand 2014-05-05 18:09:58 UTC
-- Backtrace:
Application: Konqueror (konqueror), signal: Segmentation fault
Using host libthread_db library "/lib/libthread_db.so.1".
[Current thread is 1 (Thread 0xb77bea80 (LWP 1460))]

Thread 4 (Thread 0xb4d22b40 (LWP 1462)):
#0  0xb77ed424 in __kernel_vsyscall ()
#1  0x42f81e22 in clock_gettime () from /lib/libc.so.6
#2  0x473619cd in qt_gettime() () from /lib/libQtCore.so.4
#3  0x47453c73 in QTimerInfoList::timerWait(timeval&) () from /lib/libQtCore.so.4
#4  0x4745208b in timerSourcePrepareHelper(GTimerSource*, int*) () from /lib/libQtCore.so.4
#5  0x4745213d in timerSourcePrepare(_GSource*, int*) () from /lib/libQtCore.so.4
#6  0x431c6dbf in g_main_context_prepare () from /lib/libglib-2.0.so.0
#7  0x431c7787 in g_main_context_iterate.isra.23 () from /lib/libglib-2.0.so.0
#8  0x431c79e9 in g_main_context_iteration () from /lib/libglib-2.0.so.0
#9  0x4745228f in QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () from /lib/libQtCore.so.4
#10 0x4741eea0 in QEventLoop::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () from /lib/libQtCore.so.4
#11 0x4741f231 in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () from /lib/libQtCore.so.4
#12 0x473000e5 in QThread::exec() () from /lib/libQtCore.so.4
#13 0x4730024c in QThread::run() () from /lib/libQtCore.so.4
#14 0x47302beb in QThreadPrivate::start(void*) () from /lib/libQtCore.so.4
#15 0x43049d8a in start_thread () from /lib/libpthread.so.0
#16 0x42f73a0e in clone () from /lib/libc.so.6

Thread 3 (Thread 0xb3eafb40 (LWP 1466)):
#0  0xb77ed424 in __kernel_vsyscall ()
#1  0x4304d7ac in pthread_cond_wait@@GLIBC_2.3.2 () from /lib/libpthread.so.0
#2  0x4cb2cc44 in WTF::ThreadCondition::wait(WTF::Mutex&) () from /lib/sse2/libQtWebKit.so.4
#3  0x4c7f38dc in JSC::BlockAllocator::blockFreeingThreadMain() () from /lib/sse2/libQtWebKit.so.4
#4  0x4cb0f139 in WTF::threadEntryPoint(void*) () from /lib/sse2/libQtWebKit.so.4
#5  0x4cb2c56a in WTF::wtfThreadEntryPoint(void*) () from /lib/sse2/libQtWebKit.so.4
#6  0x43049d8a in start_thread () from /lib/libpthread.so.0
#7  0x42f73a0e in clone () from /lib/libc.so.6

Thread 2 (Thread 0xb25ffb40 (LWP 1517)):
#0  0xb77ed424 in __kernel_vsyscall ()
#1  0x42f673eb in poll () from /lib/libc.so.6
#2  0x431d6f9c in g_poll () from /lib/libglib-2.0.so.0
#3  0x431c78a0 in g_main_context_iterate.isra.23 () from /lib/libglib-2.0.so.0
#4  0x431c79e9 in g_main_context_iteration () from /lib/libglib-2.0.so.0
#5  0x431c7a76 in glib_worker_main () from /lib/libglib-2.0.so.0
#6  0x431eeebb in g_thread_proxy () from /lib/libglib-2.0.so.0
#7  0x43049d8a in start_thread () from /lib/libpthread.so.0
#8  0x42f73a0e in clone () from /lib/libc.so.6

Thread 1 (Thread 0xb77bea80 (LWP 1460)):
[KCrash Handler]
#6  0x4b410610 in void QWebPagePrivate::mousePressEvent<QMouseEvent>(QMouseEvent*) () from /lib/sse2/libQtWebKit.so.4
#7  0x4b41b474 in QWebPage::event(QEvent*) () from /lib/sse2/libQtWebKit.so.4
#8  0x4b41e012 in QWebView::mousePressEvent(QMouseEvent*) () from /lib/sse2/libQtWebKit.so.4
#9  0x4a0c77a4 in KWebView::mousePressEvent(QMouseEvent*) () from /lib/libkdewebkit.so.5
#10 0x482b92dd in QWidget::event(QEvent*) () from /lib/libQtGui.so.4
#11 0x4b41f758 in QWebView::event(QEvent*) () from /lib/sse2/libQtWebKit.so.4
#12 0x4825ac04 in QApplicationPrivate::notify_helper(QObject*, QEvent*) () from /lib/libQtGui.so.4
#13 0x482638c6 in QApplication::notify(QObject*, QEvent*) () from /lib/libQtGui.so.4
#14 0x47eadec0 in KApplication::notify(QObject*, QEvent*) () from /lib/libkdeui.so.5
#15 0x474205f6 in QCoreApplication::notifyInternal(QObject*, QEvent*) () from /lib/libQtCore.so.4
#16 0x48261312 in QApplicationPrivate::sendMouseEvent(QWidget*, QMouseEvent*, QWidget*, QWidget*, QWidget**, QPointer<QWidget>&, bool) () from /lib/libQtGui.so.4
#17 0x482e6de8 in QETWidget::translateMouseEvent(_XEvent const*) () from /lib/libQtGui.so.4
#18 0x482e54d3 in QApplication::x11ProcessEvent(_XEvent*) () from /lib/libQtGui.so.4
#19 0x483111bb in x11EventSourceDispatch(_GSource*, int (*)(void*), void*) () from /lib/libQtGui.so.4
#20 0x431c7556 in g_main_context_dispatch () from /lib/libglib-2.0.so.0
#21 0x431c7920 in g_main_context_iterate.isra.23 () from /lib/libglib-2.0.so.0
#22 0x431c79e9 in g_main_context_iteration () from /lib/libglib-2.0.so.0
#23 0xaed60982 in gtk_main_iteration () from /lib/libgtk-x11-2.0.so.0
#24 0xaf60a08c in ?? () from /usr/lib/flash-plugin/libflashplayer.so
#25 0x431c8262 in g_timeout_dispatch () from /lib/libglib-2.0.so.0
#26 0x431c7556 in g_main_context_dispatch () from /lib/libglib-2.0.so.0
#27 0x431c7920 in g_main_context_iterate.isra.23 () from /lib/libglib-2.0.so.0
#28 0x431c79e9 in g_main_context_iteration () from /lib/libglib-2.0.so.0
#29 0x47452270 in QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () from /lib/libQtCore.so.4
#30 0x4831137c in QGuiEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () from /lib/libQtGui.so.4
#31 0x4741eea0 in QEventLoop::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () from /lib/libQtCore.so.4
#32 0x4741f231 in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () from /lib/libQtCore.so.4
#33 0x4742506b in QCoreApplication::exec() () from /lib/libQtCore.so.4
#34 0x48258f15 in QApplication::exec() () from /lib/libQtGui.so.4
#35 0x4651f83a in kdemain () from /lib/libkdeinit4_konqueror.so
#36 0x0804872c in main ()
Comment 2 rembrand 2014-06-13 19:25:38 UTC
It looks like this is related to the latest flash plugins and MAYBE the hardware in use.
This machine uses the ATI RADEON 3400 series with the default mesalib drivers that comes with Fedora 20. The crashes are always related to konquer/webkit mode or PyQt4 with webkit. When testing with Firefox the seg faults do not occur.

So far Linux Mint 16 (Cinnemon) inside a VirtualBox VM on the same Fedora 20 host does NOT show any problems with the PyQt4/Webkit flash plugins.

The seg faults do occur with
1) the latest Flash plugin libflashplayer.so (11.2.x.x)
2) the latest Gnash plugin  libgnashplugin.so
3) the older Flash plugin libflashplayer.so (9.x.x.x and 10.x.x.x) 

HOWEVER the crashed do much LESS occur with the libflashplayer.so 10.x

Notes: 
a) ONLY /usr/lib/mozilla/plugins is used as a plugin resource directory. All other system and home directories related to mozilla plugins are clean/empy/do not contain any kind of plugins.

b) Tests are all done with a single plugin installed each time (there where no other plugins at the time of the crashes inside /usr/lib/mozilla/plugins).

c) hardware acceleration on/of does not have any effect on stability
inside Fedora 20 KDE or the flashplugin.
Comment 3 rembrand 2014-06-13 19:39:10 UTC
The crashes do occur not after just 10 seconds but between 10 seconds and 15 minutes or so when browsing/playing movies and entering text on Youtube.
Comment 4 Dawit Alemayehu 2014-06-14 11:48:03 UTC
I cannot reproduce this crashe with KDE 4.13 and the latest adobe flash plugin on ArchLinux. Have you tried removing the flash/adobe directories?

rm -rf ~/.adobe/ ~/.macromedia/

This is especially important if you switch between the different flash plugins. Actually you only need to remove the cache, but I remove all of it from time to time. If you do not want to remove them, rename them to something else temporarily for the test.
Comment 5 rembrand 2014-06-24 12:47:52 UTC
Yes, I did remove the directories but without any result, the crashes do still occur on (/lib/libthread_db.so.1). I saw other bugs related to Google sites and it looks to me that it has to do with Java and Fedora 20 .
Comment 6 Andrew Crouthamel 2018-10-31 04:09:27 UTC
Dear Bug Submitter,

This bug has been stagnant for a long time. Could you help us out and re-test if the bug is valid in the latest version? I am setting the status to NEEDSINFO pending your response, please change the Status back to REPORTED when you respond.

Thank you for helping us make KDE software even better for everyone!
Comment 7 Bug Janitor Service 2018-11-15 10:46:36 UTC
Dear Bug Submitter,

This bug has been in NEEDSINFO status with no change for at least
15 days. Please provide the requested information as soon as
possible and set the bug status as REPORTED. Due to regular bug
tracker maintenance, if the bug is still in NEEDSINFO status with
no change in 30 days the bug will be closed as RESOLVED > WORKSFORME
due to lack of needed information.

For more information about our bug triaging procedures please read the
wiki located here:
https://community.kde.org/Guidelines_and_HOWTOs/Bug_triaging

If you have already provided the requested information, please
mark the bug as REPORTED so that the KDE team knows that the bug is
ready to be confirmed.

Thank you for helping us make KDE software even better for everyone!
Comment 8 Bug Janitor Service 2018-11-30 03:59:44 UTC
This bug has been in NEEDSINFO status with no change for at least
30 days. The bug is now closed as RESOLVED > WORKSFORME
due to lack of needed information.

For more information about our bug triaging procedures please read the
wiki located here:
https://community.kde.org/Guidelines_and_HOWTOs/Bug_triaging

Thank you for helping us make KDE software even better for everyone!