[mythtv-users] Mytharchive errors

[Harry Devine]

Paul Harrison wrote:
> Nick Rout wrote:
>   
>> On Tue, Dec 9, 2008 at 12:14 PM, Harry Devine <lifter89 at comcast.net> wrote:
>>   
>>     
>>> Nick Rout wrote:
>>>     
>>>       
>>   
>>     
>>>> as I said try the ffmpeg line from the command line and check the
>>>> output, which should give a better error message.
>>>>
>>>> midentify only identifies what codecs and formats the original file is in.
>>>> _______________________________________________
>>>>
>>>>       
>>>>         
>>> I did that last week, and it ran for 24 hours before I killed it.
>>> Should I let it run for 48 hours?  Maybe 72? ;-)
>>>     
>>>       
>> just looking at the command I see that part of it is "-i
>> /tmp/work/1/audout" - this means it is using the file
>> /tmp/work/1/audout as its input (-i) - I suspect that
>> /tmp/work/1/audout is a pipe or something [1] and there should be
>> another command filling it. If the (filling) other command isn't
>> running then the command you are running will sit around forever
>> waiting for input from /tmp/work/1/audout.
>>
>> I'd need to have a look at the whole mytharchive script to debug
>> further, but I don't have it here.
>>
>> The other option is to look at the file where mytharchive keeps its
>> logs - something like /var/lib/mytharchive/log. There are at least two
>> log files, a summary and a detailed. I think only the summary goes to
>> the screen, leaving a bigger and better log in the directory.
>>
>>   
>>     
>
> Nick is right you need to run mythtranscode to create the fifos before 
> running ffmpeg.
>
> Something like :-
> mythtranscode -p 27 -i INPUTFILE -f OUTPUTFOLDER
>
> where INPUTFILE is the name of the file you want to transcode and 
> OUTPUTFOLDER is the directory where  the fifos should be created. Then 
> run the ffmpeg command. ffmpeg reads the decoded video and audio from 
> these fifos.
>
> Not sure you've said what type of card you use to record but it sounds 
> like you've made life hard for yourself by re-encoding  the original 
> file to a nuv file. If the original recording was an mpeg2 file then 
> there may be no need to re-encode the file at all. If it does need to be 
> re-encoded then ffmpeg can handle both the decoding and encoding side 
> for mpeg2 files. It is because ffmpeg doesn't know how to decode myths 
> nuv files that we need to use mythtranscode to do the decoding which 
> complicates things somewhat.
>   
> Paul H.
>
> _______________________________________________
> mythtv-users mailing list
> mythtv-users at mythtv.org
> http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
>
>   
OK, gotcha.  That makes sense.  I know that I checked "Auto Transcode" 
when I setup the recording using the Autodetect profile.  Sounds like 
that was a mistake.  Anyhow, I'm trying to run mythtranscode now to put 
it into a better format.  From my /storage/recordings directory, I run 
the following command:

mythtranscode -p 27 -i 1023_20081130195900.nuv -f /storage/temp 
--showprogress

Its been running for roughly 30 minutes now, with the last status output 
saying "mythtranscode: 0% Completed @ 0 fps.".  Is this usually a long 
process?  Is that command not right?  And what does the 27 mean after 
the -p?  All the Wiki says is that -p specifies a profile, but I'm not 
sure what 27 means.

Thanks for the help!
Harry