[mythtv-users] Recordings from nonexistent channels

[Michael T. Dean]

On 08/06/2012 10:02 AM, Jan Ceuleers wrote:
> On 07/30/2012 09:51 PM, Jan Ceuleers wrote:
>> Hi there
>>
>> I'm on 0.24-fixes.
>>
>> Occasionally (but too often to my liking) recordings
>> fail. By "fail" I mean that something was recorded
>> (i.e. good audio and video) but that it does not
>> correspond with the metadate (i.e. with the program
>> that I expected to be recorded).
>>
>> Looking at the backed logs, I see that the recording
>> in question was recorded on a channel that does not
>> exist. Example:
>>
>> Finished recording 	Anvil: the Story of Anvil:
>> channel 2019
>>
>> Where channel 19 does not exist (i.e. select * from
>> channel where chanid=2019; returns the empty set).
>>
>> Any idea how to troubleshoot this?
> I have a theory.
>
> These recordings seem to be happening from channels for
> which I have metadata but no valid tuning data. So when
> another-one happened overnight I went back to the
> listings to see which channel the recording's metadata
> suggested it had come from.
>
> I do grab metadata (using uk_rt) for more channels than
> I actually subscribe to. These channels also exist in
> the database, but have been rendered invisible (by
> clearing the "visible" flag in Mythweb).
>
> I thought that clearing that flag caused the channel to
> not only disappear from the program listings but also
> caused it to be removed from consideration for
> recordings. I now suspect that this may not hold for
> power rules.
>
> So I will now change my grabbers's configuration files
> to only grab metadata for channels I actually subscribe
> to, and I'm sure that this will solve the problem.

Jan, this sounds like a good theory.  We do have some plans for changes 
to make the visible flag work more like people expect it to work 
(basically turning it into a "disabled" flag), which could actually 
allow you to use it like you have been trying.  The behavior you're 
seeing makes me think we should look at making that change sooner rather 
than later.

I'm sorry I hadn't yet gotten a chance to look over your database to 
come up with a theory--I was on a trip until today, so didn't get time 
to look it over.  Please let me know if you see the issue, again, after 
getting rid of the extra channels.

Mike