rhythmbox segfault in gdk_window_set_transient_for

Thanks for your bug. Could you install libgtk2.0-0-dbg and try to get a backtrace with it?

Thanks for that debug package. Here is a new backtrace:

#0 0xb75c984b in IA__gdk_window_set_transient_for (window=0x871c680, parent=0x8467860) at gdkwindow-x11.c:2913
#1 0xb7854b60 in gtk_window_transient_parent_realized (parent=0x82ce800, window=0x8627090) at gtkwindow.c:1831
#2 0xb7858566 in IA__gtk_window_set_transient_for (window=0x84ade90, parent=0x82ce800) at gtkwindow.c:1946
#3 0xb7802c19 in gtk_tooltips_draw_tips (tooltips=0x82e0270) at gtktooltips.c:443
#4 0xb780300b in gtk_tooltips_timeout (data=0x82e0270) at gtktooltips.c:458
#5 0xb6ff9e46 in g_source_get_current_time () from /usr/lib/libglib-2.0.so.0
#6 0xb6ff9872 in g_main_context_dispatch () from /usr/lib/libglib-2.0.so.0
#7 0xb6ffc84f in g_main_context_check () from /usr/lib/libglib-2.0.so.0
#8 0xb6ffcbf9 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0
#9 0xb772d464 in IA__gtk_main () at gtkmain.c:1024
#10 0x0806ce84 in main ()

(I tried twice to attach the full gdb output, launchpad is not behaving.)

Do you still have that problem? What window manager do you use? Could you attach a complete crash file to the bug maybe?

Sorry, I haven't been able to reproduce. This is standard Ubuntu with Gnome and metacity.

What is the gdb file if you have not been able to reproduce? The backtrace doesn't look wrong, could you get a valgrind log as described on https://wiki.ubuntu.com/Valgrind?

The gdb file was from running gdb on the old coredump, but after installing -dbg package. I have tried to reproduce but no luck. It might be a race condition / timing issue, I have an old laptop where I am able to do some clicking faster than gtk would be used to. I filed the bug anyway, in case other people would have seen something similar. I can provide the core dump (later) if you like, but maybe we should rather wait for me or others to reproduce.

Could you rather try getting the valgrind log as described before?

Sure, I will try (after vacations, i.e. next year). Is the valgrind log useful even without reproducing the crash?

it might be useful, debug log can point incorrect uses which can lead to a crash

No reply, bug closed. Feel free to reopen with the asked informations if you still get that problem