firefox crash while using moonlight

Bug #367279 reported by Andrea Bravetti on 2009-04-26
28
This bug affects 5 people
Affects Status Importance Assigned to Milestone
alsa-plugins
New
Unknown
alsa-plugins (Ubuntu)
Undecided
Unassigned

Bug Description

Firefox crash while using moonlight to see a stream.

To reproduce this problem go there and wait:
http://www.rai.tv/dl/RaiTV/diretta.html?cid=ContentSet-37f0f5d4-af33-48df-9281-2d46214c1b5d

This is the log:

andrea@kubra:~$ firefox
DownloaderRequest: http://livestream.rai.it.edgestreams.net/reflector:64084?auth=daEa4bjbtcqdDbLaIa6b8dPcgbzaXbtaBc9-bj9ewx-b4-GkAvAGt&aifp=V001
DownloaderRequest: http://livestream.rai.it.edgestreams.net/reflector:64084?auth=daEa4bjbtcqdDbLaIa6b8dPcgbzaXbtaBc9-bj9ewx-b4-GkAvAGt&aifp=V001
[wmv3 @ 0xad36c00]Extra data: 8 bits left, value: 0
AudioPlayer: Using PulseAudio.
Moonlight: Connection failure while trying to connect to pulseaudio daemon: Connection refused
AudioPlayer: Using Alsa.
Segmentation fault

I'm on Kubuntu Jaunty fully upgraded at 2009/04/26.

I don't know what is the "connection failure" but it happen as soon as the video start, and it is not related to the crash (or it seems so...).
edit: it may be because I don't use pulseaudio...

I'm on kubuntu jaunty fully upgraded at 2009/04/26.

It may be related to some codec and not to moonlight, but rigth now I don't know how to locate the exact cause.

Before the crash all work properly for some time.

description: updated

some time it may take an hour to crash, but it will crash...

Jo Shields (directhex) wrote :

This site is region-locked to Italians only, which makes verifying and diagnosing it very difficult

Jo Shields (directhex) wrote :

Are you using the built-in ffmpeg codecs, or the Microsoft codecs? (right-click on the Moonlight control to get the install option). If the former, try the latter. If it doesn't crash with the MS codecs, it's an FFmpeg bug.

I'm using ffmpeg rigth now, I will try with the ms codec...

> This site is region-locked to Italians only, which makes verifying and diagnosing it very difficult

I can setup a proxy for you if you can debug it...

Now I'm testing it with the ms codec...

it crash in the same way even with the ms codec.

<OT>
and I tainted one of my linux box... should I reinstall it from scratch? No, I'm joking, but how to remove this codec?
removing .mozilla/plugins/moonlight/silverlight-media-pack-linux-x86-5-1.so is enought?
</OT>

Jo Shields (directhex) wrote :

Yes, deleting that file is enough.

Can you follow the instructions on https://wiki.ubuntu.com/MozillaTeam/Bugs#Crashes to obtain a backtrace? You'll want the libmoon-dbgsym, moonlight-plugin-core-dbgsym and moonlight-plugin-mozilla-dbgsym packages as well as those named. I can then present the info to upstream, who can help find a fix

I followed the istructions at https://wiki.ubuntu.com/MozillaTeam/Bugs#Crashes and gdb showed this:

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0xb7dba6d0 (LWP 7803)]
0xae101b73 in alsa_lib_resampler_process_int () from /usr/lib/alsa-lib/libasound_module_rate_speexrate.so

so, I installed libasound2-plugins-dbgsym and that message become:

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0xb7c6c6d0 (LWP 8219)]
0xadf2ab73 in alsa_lib_resampler_process_int (st=0x9824f80, channel_index=0, in=0xbd0abc58, in_len=0xbf94a830,
    out=0x3c28f927, out_len=0xbf94a828) at resample.c:915
915 resample.c: No such file or directory.
        in resample.c

Attached gdb-firefox1.txt (normal) and gdb-firefox2.txt (with libasound2-plugins-dbgsym).

So, I installed source for libasound2-plugins and the offending line seems to be this:

   for (i=0;i<*out_len;i++)
      out[i*st->out_stride] = WORD2INT(y[i]);

If I'm not wrong the bug should be assigned to libasound2-plugins.

affects: moon (Ubuntu) → alsa-plugins (Ubuntu)

I was wrong...
If this time I'm not wrong the bug should be assigned to alsa-plugins (source of libasound2-plugins)....

Jorge Castro (jorge) on 2009-06-09
Changed in alsa-plugins:
importance: Undecided → Unknown
status: New → Unknown
Daniel T Chen (crimsun) wrote :

Can you reproduce this symptom in current lucid?

Daniel T Chen (crimsun) on 2010-03-07
Changed in alsa-plugins (Ubuntu):
status: New → Triaged
Changed in alsa-plugins:
status: Unknown → New
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.