Bug 318816 – Doesn't like it when mDNSResponder goes away

After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.

Bug 318816 - Doesn't like it when mDNSResponder goes away


Summary:	Doesn't like it when mDNSResponder goes away


Status:	RESOLVED INVALID

Product:	rhythmbox
Classification:	Other
Component:	DAAP
Version:	HEAD
Hardware:	Other Linux

Importance:	Normal normal
Target Milestone:	---
Assigned To:	RhythmBox Maintainers
QA Contact:	RhythmBox Maintainers

URL:
Whiteboard:

Depends on:	319305
Blocks:

Reported:	2005-10-13 22:20 UTC by Bastien Nocera
Modified:	2008-08-14 13:16 UTC

See Also:
GNOME target:	---
GNOME version:	---

Description Bastien Nocera 2005-10-13 22:20:41 UTC

1. start Rhythmbox with DAAP enabled, and mDNSResponder running
2. stop mDNSResponder
3. Rhythmbox will now eat all your CPU
4. Exit Rhythmbox, and see a segfault:

+ Trace 63526

#0 sw_corby_channel_start_request
from /usr//lib/libhowl.so.0
#1 sw_corby_object_start_request
from /usr//lib/libhowl.so.0
#2 sw_mdns_stub_cancel
from /usr//lib/libhowl.so.0
#3 sw_discovery_cancel
from /usr//lib/libhowl.so.0
#4 rb_daap_mdns_resolve_cancel
at rb-daap-mdns.c line 323
#5 stop_resolver
at rb-daap-source.c line 400
#6 g_hash_table_destroy
from /usr//lib/libglib-2.0.so.0
#7 stop_browsing
at rb-daap-source.c line 438
#8 rb_daap_sources_shutdown
at rb-daap-source.c line 506
#9 rb_shell_quit
at rb-shell.c line 1854
#10 g_cclosure_marshal_VOID__VOID
from /usr//lib/libgobject-2.0.so.0
#11 g_closure_invoke
from /usr//lib/libgobject-2.0.so.0
#12 g_signal_stop_emission
from /usr//lib/libgobject-2.0.so.0
#13 g_signal_emit_valist
from /usr//lib/libgobject-2.0.so.0
#14 g_signal_emit
from /usr//lib/libgobject-2.0.so.0
#15 gtk_action_create_menu_item
from /usr//lib/libgtk-x11-2.0.so.0
#16 gtk_action_activate
from /usr//lib/libgtk-x11-2.0.so.0
#17 g_cclosure_marshal_VOID__VOID
from /usr//lib/libgobject-2.0.so.0
#18 g_closure_invoke
from /usr//lib/libgobject-2.0.so.0
#19 g_signal_stop_emission
from /usr//lib/libgobject-2.0.so.0
#20 g_signal_emit_valist
from /usr//lib/libgobject-2.0.so.0
#21 g_signal_emit
from /usr//lib/libgobject-2.0.so.0
#22 gtk_widget_activate
from /usr//lib/libgtk-x11-2.0.so.0
#23 gtk_menu_shell_activate_item
from /usr//lib/libgtk-x11-2.0.so.0
#24 gtk_menu_shell_activate_item
from /usr//lib/libgtk-x11-2.0.so.0
#25 gtk_menu_reorder_child
from /usr//lib/libgtk-x11-2.0.so.0
#26 gtk_marshal_VOID__UINT_STRING
from /usr//lib/libgtk-x11-2.0.so.0
#27 g_cclosure_new_swap
from /usr//lib/libgobject-2.0.so.0
#28 g_closure_invoke
from /usr//lib/libgobject-2.0.so.0
#29 g_signal_stop_emission
from /usr//lib/libgobject-2.0.so.0
#30 g_signal_emit_valist
from /usr//lib/libgobject-2.0.so.0
#31 g_signal_emit
from /usr//lib/libgobject-2.0.so.0
#32 gtk_widget_activate
from /usr//lib/libgtk-x11-2.0.so.0
#33 gtk_propagate_event
from /usr//lib/libgtk-x11-2.0.so.0
#34 gtk_main_do_event
from /usr//lib/libgtk-x11-2.0.so.0
#35 gdk_screen_get_setting
from /usr//lib/libgdk-x11-2.0.so.0
#36 g_main_context_dispatch
from /usr//lib/libglib-2.0.so.0
#37 g_main_context_check
from /usr//lib/libglib-2.0.so.0
#38 g_main_loop_run
from /usr//lib/libglib-2.0.so.0
#39 gtk_main
from /usr//lib/libgtk-x11-2.0.so.0
#40 main
at main.c line 381

Comment 1 Ernst Sjöstrand 2005-10-18 10:36:05 UTC

I see something like this with avahi also.

Comment 2 Bastien Nocera 2005-10-27 10:26:21 UTC

That might help the CPU spinning:
http://j.kaolin.hn.org/rhythmbox/libsoup-hangup-fix.patch
Also in bug #319305

Comment 3 Alex Lancaster 2006-01-26 15:12:32 UTC

Mark as a DAAP issue.

Comment 4 William Jon McCann 2006-02-14 22:55:00 UTC

Do you still see this with head?  DAAP has been reworked quite a bit.  I don't see it with avahi.

Comment 5 Bastien Nocera 2006-03-09 13:35:43 UTC

It's still well broken, I'm pretty sure it's not RB's fault, but in the meanwhile, I don't have any proofs of that...

Comment 6 Hubert Figuiere (:hub) 2008-08-14 13:10:11 UTC

doesn't RB use Avahi now? I haven't had crashes in a good while when mDNS goes down, and the above stack trace isn't in Avahi.

Comment 7 Jonathan Matthew 2008-08-14 13:16:48 UTC

I removed the howl code recently, after it had been more or less unused for a couple of years.  If there's an equivalent problem with avahi, then we should already have an open bug about it.