reassign 448621 python-gst0.10 tag 448621 + fixed-upstream thanks Le vendredi 02 novembre 2007 à 19:08 +0100, Julien Valroff a écrit : > Le vendredi 02 novembre 2007 à 18:55 +0100, Josselin Mouette a écrit : > > Hi, > > > > Le vendredi 02 novembre 2007 à 18:51 +0100, Julien Valroff a écrit : > > > Hi Josselin, > > > > > > Le vendredi 02 novembre 2007 à 17:40 +0100, Josselin Mouette a écrit : > > > > This could well be a pygobject bug indeed, but I think we need a better > > > > backtrace. Could you by any chance rebuild pygobject with debugging > > > > symbols enabled and obtain a full backtrace (thread apply bt full)? > > > > > > I am not sure I have done what you want... > > > Does the following help? > [...] > > > > Here you need to type "thread apply all bt full". > That wasn't clear for me when I have first read the page you refer to... > I had typed this before ;-) > > > More information on how to rebuild the package and provide the necessary > > information: http://wiki.debian.org/HowToGetABacktrace > > Please find the output attached.
OK, turns out this is a bug in pygst which frees all resources in atexit, which in turn goes back in the python interpreter, leading to a locking issue. It is already fixed in the GStreamer CVS: http://webcvs.freedesktop.org/gstreamer/gst-python/gst/gstmodule.c?r1=1.46&r2=1.47 Cheers, -- .''`. : :' : We are debian.org. Lower your prices, surrender your code. `. `' We will add your hardware and software distinctiveness to `- our own. Resistance is futile.
signature.asc
Description: Ceci est une partie de message numériquement signée