suddenly >88 processes

[Mark Kretschmann]

Looking at the backtraces below, I would not be surprised if it was caused by 
our use of debug() statements in the crashhandler, causing a kind of a 
recursive crash. I've had something similar happen the other day, although 
only with one level of recursion. Amarok crashed, and then crashed again, 
invoking two instances of KMail for backtrace sending.

How this exactly happens we might never know, but Alex went ahead and removed 
the debug() statements from our crashhandler, so we should be save now.


On Friday 30 June 2006 03:23, paulc2 at optonline.net wrote:
> The backtrace below is described as one of the forked processes, and it
> definitely isnt in the engine.
>
> Quoting Jeff Mitchell <kde-dev at emailgoeshere.com>:
> > I've seen this behavior on Helix before.  In fact, I can easily
> > reproduce it...just load up a song with Helix engine, and then start
> > seeking back and forth in the song rapidly.  Forkbombs ahoy.  Never
> > got a good backtrace because half the time I ended up just killing X.
> >
> > Quoting paulc2 at optonline.net:
> >> just noting that this trace does not seem to be engine related:
> >>
> >> The forked ones had backtraces that looked like this:
> >>
> >> #0  0xffffe410 in __kernel_vsyscall ()
> >> #1  0x425d0fbb in __waitpid_nocancel () from /lib/libpthread.so.0
> >> #2  0x0804e8c5 in amaroK::Crash::crashHandler ()
> >> #3  <signal handler called>
> >> #4  0x42da2802 in QGListIterator::QGListIterator () from
> >> /usr/qt/3/lib/libqt-mt.so.3
> >> #5  0x42ae7e62 in QObject::child () from /usr/qt/3/lib/libqt-mt.so.3
> >> #6  0x0804f3a5 in Debug::indent ()
> >> #7  0x0804dae4 in QGList::count ()
> >> #8  0x0804dc53 in amaroK::Crash::crashHandler ()
> >> #9  <signal handler called>
> >> #10 0x42da2802 in QGListIterator::QGListIterator () from
> >> /usr/qt/3/lib/libqt-mt.so.3
> >> #11 0x42ae7e62 in QObject::child () from /usr/qt/3/lib/libqt-mt.so.3
> >> #12 0x0804f3a5 in Debug::indent ()
> >> #13 0x0804dae4 in QGList::count ()
> >> #14 0x0804dc53 in amaroK::Crash::crashHandler ()
> >> #15 <signal handler called>
> >> #16 0x42da2802 in QGListIterator::QGListIterator () from
> >> /usr/qt/3/lib/libqt-mt.so.3
> >> #17 0x42ae7e62 in QObject::child () from /usr/qt/3/lib/libqt-mt.so.3
> >> #18 0x0804f3a5 in Debug::indent ()
> >> #19 0x0804dae4 in QGList::count ()
> >> #20 0x0804dc53 in amaroK::Crash::crashHandler ()
> >> #21 <signal handler called>
> >> #22 0x42da2802 in QGListIterator::QGListIterator () from
> >> /usr/qt/3/lib/libqt-mt.so.3
> >> #23 0x42ae7e62 in QObject::child () from /usr/qt/3/lib/libqt-mt.so.3
> >> #24 0x0804f3a5 in Debug::indent ()
> >> #25 0x0804dae4 in QGList::count ()
> >> #26 0x0804dc53 in amaroK::Crash::crashHandler ()
> >> #27 <signal handler called>
> >>
> >> Quoting Colin Brace <cb at lim.nl>:
> >>> On 6/28/06, Martin Aumueller <aumueller at uni-koeln.de> wrote:
> >>>> Could this be related to this bug?
> >>>> https://bugs.kde.org/show_bug.cgi?id=126397
> >>>
> >>> Looks very similar. Just before it happened, I also heard the the
> >>> player start "skipping" on a track I was playing.
> >>>
> >>> I am also using the helix engine on my FC5 system.

-- 
Mark