Transmission freezes when adding second magnet in Ubuntu 19.04

[Toan Nguyen]

Step to reproduce:
1. Open browser
2. Click a magnet link
3. Choose to use transmission
4. Choose folder
5. OK
6. Open another magnet link
7. Choose folder
--> It freezes

[Sebastien Bacher (seb128)]

Thank you for your bug report. What is a magnet? Could you add your journalctl log to the bug after getting the issue?

[Toan Nguyen (dtoan96)]

Magnet link is a kind of torrent without the need of downloading a torrent file (you just add link and it starts downloading).
I tried to use journalctl and this seems to be where the problem is:

Thg 4 24 19:13:32 DINGO transmission-gt[3579]: g_object_set_data: assertion 'G_IS_OBJECT (object)' failed
Thg 4 24 19:13:32 DINGO transmission-gt[3579]: gdk_window_get_window_type: assertion 'GDK_IS_WINDOW (window)' failed
Thg 4 24 19:13:32 DINGO transmission-gt[3579]: _gtk_widget_captured_event: assertion 'WIDGET_REALIZED_FOR_EVENT (widget, event)' failed
Thg 4 24 19:13:32 DINGO transmission-gt[3579]: _gtk_widget_captured_event: assertion 'WIDGET_REALIZED_FOR_EVENT (widget, event)' failed
Thg 4 24 19:13:32 DINGO transmission-gt[3579]: _gtk_widget_captured_event: assertion 'WIDGET_REALIZED_FOR_EVENT (widget, event)' failed

[Sebastien Bacher (seb128)]

Could you give an url example to test the issue?

[Toan Nguyen (dtoan96)]

The issue seems to only appear when directly clicking link from browser. You can go to https://nyaa.si/ in the "Link" column, there's U-like symbols, those are magnet link. Just click it.

[Sebastien Bacher (seb128)]

Thanks

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Sebastien Bacher (seb128)]

Using the steps described it doesn't hang there. Could you get a backtrace using the instructions on https://wiki.ubuntu.com/Backtrace#Already_running_programs ?

[Toan Nguyen (dtoan96)]

This is the backtrace of the bug.

[Toan Nguyen (dtoan96)]

This is the video of the whole process in case you need it.

[Sebastien Bacher (seb128)]

Thanks, it's a bit weird that ctrl-C in gdb doesn't work and that you don't get a prompt back after killing the process. Could you try by exiting transmission and doing
$ gdb transmission-gtk
(gdb) r <enter>
... get the issue
(gdb) <Ctrl-C>
(gdb) t a a bt

[Toan Nguyen (dtoan96)]

This is the output after (gdb) t a a bt:

Thread 13 (Thread 0x7fffcdffb700 (LWP 7985)):
#0 0x00007ffff6bc4729 in __GI___poll (fds=0x555555f4b270, nfds=1, timeout=-1) at ../sysdeps/unix/sysv/linux/poll.c:29
#1 0x00007ffff7426cb6 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#2 0x00007ffff7426ddc in g_main_context_iteration () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#3 0x00007ffff7fc5ffd in ?? () from /usr/lib/x86_64-linux-gnu/gio/modules/libdconfsettings.so
#4 0x00007ffff744f98d in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#5 0x00007ffff6ca7182 in start_thread (arg=<optimized out>) at pthread_create.c:486
#6 0x00007ffff6bd0b1f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95

Thread 11 (Thread 0x7fffceffd700 (LWP 7983)):
#0 syscall () at ../sysdeps/unix/sysv/linux/x86_64/syscall.S:38
#1 0x00007ffff7471b9a in g_cond_wait_until () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#2 0x00007ffff73f80c1 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#3 0x00007ffff73f8681 in g_async_queue_timeout_pop () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#4 0x00007ffff74503f1 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
--Type <RET> for more, q to quit, c to continue without paging--
#5 0x00007ffff744f98d in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#6 0x00007ffff6ca7182 in start_thread (arg=<optimized out>) at pthread_create.c:486
#7 0x00007ffff6bd0b1f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95

Thread 8 (Thread 0x7fffcffff700 (LWP 7972)):
#0 0x00007ffff6bc4729 in __GI___poll (fds=0x7fffc8007080, nfds=3, timeout=-1) at ../sysdeps/unix/sysv/linux/poll.c:29
#1 0x00007fffe08359f1 in ?? () from /lib/x86_64-linux-gnu/libpulse.so.0
#2 0x00007fffe0827260 in pa_mainloop_poll () from /lib/x86_64-linux-gnu/libpulse.so.0
#3 0x00007fffe08278ae in pa_mainloop_iterate () from /lib/x86_64-linux-gnu/libpulse.so.0
#4 0x00007fffe0827960 in pa_mainloop_run () from /lib/x86_64-linux-gnu/libpulse.so.0
#5 0x00007fffe0835939 in ?? () from /lib/x86_64-linux-gnu/libpulse.so.0
#6 0x00007fffe07d5788 in ?? () from /usr/lib/x86_64-linux-gnu/pulseaudio/libpulsecommon-12.2.so
#7 0x00007ffff6ca7182 in start_thread (arg=<optimized out>) at pthread_create.c:486
#8 0x00007ffff6bd0b1f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95
--Type <RET> for more, q to quit, c to continue without paging--

Thread 6 (Thread 0x7fffe3fff700 (LWP 7969)):
#0 0x00007ffff6cb1440 in __GI___nanosleep (requested_time=0x7fffe3ffe800, remaining=0x0) at ../sysdeps/unix/sysv/linux/nanosleep.c:28
#1 0x00005555555b5c66 in tr_wait_msec ()
#2 0x00005555555b2772 in ?? ()
#3 0x000055555559f63a in ?? ()
#4 0x00007ffff6ca7182 in start_thread (arg=<optimized out>) at pthread_create.c:486
#5 0x00007ffff6bd0b1f in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:95

Thread 5 (Thread 0x7ffff0b4f700 (LWP 7968)):
#0 0x00007ffff6bd0e3f in epoll_wait (epfd=12, events=0x7fffdc000e20, maxevents=32, timeout=39) at ../sysdeps/unix/sysv/linux/epoll_wait.c:30
#1 0x00007ffff71b02c5 in ?? () from /lib/x86_64-linux-gnu/libevent-2.1.so.6
#2 0x00007ffff71a6114 in event_base_loop () from /lib/x86_64-linux-gnu/libevent-2.1.so.6
#3 ...

[Sebastien Bacher (seb128)]

The backtrace doesn't have anything obviously wrong (it lacks debug symbols also but unsure it would make a difference), setting back to "New" until someone has time to poke at it more, unsure why it doesn't happen here and is so consitent for you though

[Sebastien Bacher (seb128)]

(do you use any non default option maybe?)

[Toan Nguyen (dtoan96)]

I don't use any non default option. I remember this problem suddenly happened to me one day during the time of ubuntu 19.04 beta and it's still there when I freshly installed ubuntu 19.04 stable.

[Sebastien Bacher (seb128)]

Could you maybe also report it to the upstream code writers on https://github.com/transmission/transmission ? They might have a better clue about the issue