[mythtv-users] Occasional 376 B recordings from Ceton card

Erik Jensen eriksjunk at laurelwoodart.com
Tue Oct 18 22:46:13 UTC 2011

On Tue, Oct 18, 2011 at 4:11 AM, Ronald Frazier <ron at ronfrazier.net> wrote:
> On Sun, Oct 16, 2011 at 2:15 AM, Erik Jensen
> <eriksjunk at laurelwoodart.com> wrote:
>> I am running 0.24-fixes with the latest Ceton patch, and I am getting
>> occasional 376 byte recordings. Sometimes it only causes me to miss
>> the one recording, but other times the backend completely stops
>> working until I notice and restart it, causing it to miss several
>> recordings. During the time it is non-responsive, there is constant
>> activity on the recording drive. Any ideas what the problem could be?
> Are you positive this is a channel that works and is marked Copy
> Freely? Can you view this channel with myth using livetv? Or how about
> when feeding the video directly into mplayer using the instructions on
> ceton's website:
> http://www.cetoncorp.com/infinitv/support/linux.php

Yes, they are for sure copy freely. I have caught a failed recording
immediately once or twice, and the status page said the channel is
copy freely. Also, when I tuned to the channel in livetv a minute or
two after the show started, it worked fine. Finally, I have
back-to-back recordings record on different tuners to allow three
minutes of overlap, and sometimes the second recording will fail even
though the first successfully makes it to the end.

> Assuming the channel is otherwise fine, 376 bytes is exactly the size
> of 2 mpeg-ts packets. I'm guessing those packets are a PAT and a PMT.
> It sounds like the Ceton card is feeding enough data for myth to get
> the PAT/PMT but nothing more. I'm thinking it's a case of the channel
> being CCI restricted. My cable provider doesn't actually restrict
> anything in the package I subscribe to, so I have no idea how the card
> behaves when that happens, but it seems like a likely scenario. Even
> if the channel is normally viewable, it could be a case of a temporary
> CCI restriction on the channel. Although CCI flags are typically set
> on a per-channel basis, they can in theory change from program to
> program. I don't know of any providers that do that, but it's
> possible. Another possibility is that the CCI flags are temporarily
> disappearing, and a CGMS flag is going into effect in it's absence. I
> know I've seen this happen a few times, and I've no idea why. When
> running my CCI scanner tool, it would come across a channel marked
> with Copy Generation Management System - Analog: "One Copy" (0x02),
> but then when I'd recheck the channel a minute later, it would by back
> to Copy Control Information: "Copy Free" (00).

These channels appear to consistently be "Copy Free". Is it possible
that for some reason the card doesn't get the CCI info "soon enough"
and some part of the pipeline could be timing out? I will try to see
what happens if I try to record from a CCI protected channel.

> If you can't figure out anything, you could try emailing me the 376
> byte recording and I can take a peek at it. Also try sending me a copy
> of the myth backend log (please strip it down so that it only covers a
> couple minutes before and after the recording was scheduled to start).
> However, in order for this log to be useful to me, mythbackend will
> need to be running with the "-v general,important,channel,record"
> flags.

I'll make a copy of the failed recording next time I see the problem,
and I'll also see about adding those options to my init script. Thank
you for your help.

> ---
> Ron Frazier
> _______________________________________________
> mythtv-users mailing list
> mythtv-users at mythtv.org
> http://www.mythtv.org/mailman/listinfo/mythtv-users

More information about the mythtv-users mailing list