[mythtv-users] Firewire issues w/ 0.21 in Mythbuntu 8.10

[kanetse@gmail.com]

On Mon, Nov 24, 2008 at 6:09 PM, Harry Devine <lifter89 at comcast.net> wrote:
> OK, so I just setup my firewire connection between my Motorola DCT-3412
> STB and my Myth box (Mythbuntu 8.10, obviously).  I have the card setup,
> the lineups setup, and can tune the channels.  Everything seems to look
> OK via plugreport too.  But when I just tried to start a recording on an
> HD channel only available via Firewire, and I get the following errors
> in my backend log:
>
> 2008-11-24 21:00:17.491 LFireDev(00159AFFFE163F4A): SignalReset(173->174)
> 2008-11-24 21:00:17.506 LFireDev(00159AFFFE163F4A):
> SignalReset(173->174): Updating device list -- begin
> 2008-11-24 21:00:17.807 LFireDev(00159AFFFE163F4A):
> SignalReset(173->174): Updating device list -- end
> 2008-11-24 21:00:17.859 LFireDev(00159AFFFE163F4A), Warning: No Input in
> ...
>
> I have also tried using test-mpeg2 and was able to record an mpg file,
> but its very jumpy and choppy via mplayer.  I tried opening that mpg
> file using Windows Media Player, but it complains about the codec used
> on it and won't play it.
>
> Does anyone have any ideas on what those errors mean, and how to get
> Firewire working?  I did find this article:
> https://help.ubuntu.com/community/MythTV_Firewire, but that seems to be
> geared towards myth 0.20, and when I look at my backend startup script
> it has some firewire priming using mythprime already built into it.  So
> I'm at a bit of a loss here.
>

I have this error message appearing my logs as well, although my
recordings seem to be unaffected.  From this thread, it looks like
it's a problem in the kernel:

http://www.gossamer-threads.com/lists/mythtv/users/323501

... but I don't think Ubuntu 8.10 uses a kernel that old (< 2.6.23).
That's about the only discussion I can see on this particular problem.

Personally, I'm using CentOS 5.2 (2.6.18); but have been considering
upgrading to another distro with a more recent kernel to avoid this
problem, as I believe it is causing the occasional recording to become
truncated.

So keep us updated if you manage to get rid of this problem.