Created attachment 588561 [details] System log of pulseaudio and bluetooth messages. Shows startup, some volume adjustment, and crash. Description of problem: Pulseaudio crashes when enabling a bluetooth headset. Version-Release number of selected component (if applicable): alsa-plugins-pulseaudio-1.0.25-3.fc17.i686 blueman-1.23-2.fc17.i686 bluez-4.99-1.fc17.i686 bluez-alsa-4.99-1.fc17.i686 bluez-compat-4.99-1.fc17.i686 bluez-cups-4.99-1.fc17.i686 bluez-debuginfo-4.99-1.fc17.i686 bluez-gstreamer-4.99-1.fc17.i686 bluez-hcidump-2.2-2.fc17.i686 bluez-hcidump-debuginfo-2.2-2.fc17.i686 bluez-libs-4.99-1.fc17.i686 bluez-libs-devel-4.99-1.fc17.i686 gnome-bluetooth-3.4.0-2.fc17.i686 gnome-bluetooth-libs-3.4.0-2.fc17.i686 pulseaudio-1.1-9.fc17.i686 pulseaudio-gdm-hooks-1.1-9.fc17.i686 pulseaudio-libs-1.1-9.fc17.i686 pulseaudio-libs-glib2-1.1-9.fc17.i686 pulseaudio-module-bluetooth-1.1-9.fc17.i686 pulseaudio-module-gconf-1.1-9.fc17.i686 pulseaudio-module-x11-1.1-9.fc17.i686 pulseaudio-utils-1.1-9.fc17.i686 wine-pulseaudio-1.5.3-1.fc17.i686 How reproducible: Completely reproducible. Segfaults every time. Steps to Reproduce: 1. Fresh session, observe sound services working properly using built-in sound hardware. 2. Use gnome bluetooth tools to pair / connect to paired headphones. 3. Sound-related services are gone. Observe segfault message in logs. Actual results: kernel: [ 171.585363] pulseaudio[1888]: segfault at 9000000 ip 42d70ad6 sp bf9ace44 error 4 in libc-2.15.so[42cee000+1a8000] Expected results: Beautiful music in my headphones. :) Additional info: - Same hardware was working fine in Fedora 16. - I enabled debug logging in pulseaudio, then I grabbed a log snapshot of the issue. (Attached.)
Can you try the workaround suggested here: https://bugzilla.redhat.com/show_bug.cgi?id=827629#c14
Brendan, thanks so much for pointing this out! I have tried the workaround and it helped me. I have posted a comment on bug #827629 to confirm: https://bugzilla.redhat.com/show_bug.cgi?id=827629#c19 I noticed that other reporters of this bug are using 32 bit (i686) systems as well. For myself, I have two Fedora 17 systems, one x86_64 and one i686. Only the i686 system has been running into this issue.
Good news! *** This bug has been marked as a duplicate of bug 827629 ***