[mythtvnz] Help-can't get grabber and mythfilldatabase to work right

[Stephen Worthington]

On Sun, 28 Apr 2013 16:59:09 +1200, you wrote:

>On Sun, 28 Apr 2013 14:01:22 +1200, Stephen Worthington  
><stephen_agent at jsw.gen.nz> wrote:
>
>> On Sun, 28 Apr 2013 11:24:39 +1200, you wrote:
>>
>>>
>>>>>
>>>>> You can try running the EPG update manually.  Run a terminal and login
>>>>> as the user that runs mythfrontend.  Download the EPG file with wget
>>>>> and uncompress it with gzip -d.  Then run:
>>>>>
>>>>>   mythfilldatabase --help
>>>>>
>>>>> to see all the options for running it.  I think this is what is
>>>>> needed:
>>>>>
>>>>>   mythfilldatabase --update --file --sourceid 1 --xmlfile <path to the
>>>>> xml file>
>>>>>
>>>>> It is a while and two MythTV versions since I last had to do this, so
>>>>> the options have changed a bit.  If your sourceid is not 1, change
>>>>> that to the correct value.  When run like that, I think the log will
>>>>> go to the terminal, so you might want to add:
>>>>>
>>>>>   2>&1 >mythfilldatabase.log
>>>>>
>>>>> and possibly:
>>>>>
>>>>>   --loglevel debug
>>>>>
>>>>> Does that work?  Are there any errors?
>>>>>
>>>> That works! no errors that I can see and 8 days of data in the
>>>> schedule.System info page tags the update as successful.
>>>> -Paul
>>>
>>> So what is different between the grabber getting the data and me doing  
>>> it
>>> manually?
>>> In what folder does the grabber create the EPG file?
>>> I don't appear to get any file not found errors in the mythfilldatabase
>>> terminal output.
>>>
>>> -Paul
>>
>> The tv_grab_nz.py grabber does not download the EPG data to a file -
>> it reads it directly into a Python variable in memory.  It processes
>> that data and normally writes it out to stdout to pass it on to the
>> mythfilldatabase process that called tv_grab_nz.py.  There seems to be
>> an option to get it to write the data to a file instead of to stdout,
>> and another option ("cache") to get it to write the data to a file and
>> to stdout.  But I have never used either option.
>>
>> In fact, I do not use tv_grab_nz.py in the standard manner at all
>> myself.  I have to generate EPG for my Sky channels, so I actually run
>> a script that uses wget to get the Freeview EPG and also runs epgsnoop
>> to get the Sky EPG.  It stores both in /var/www/epg so that other PCs
>> that do TV recording on my network can download the EPG files from the
>> web server on my MythTV box, and then runs mythfilldatabase.
>> Mythfilldatabase runs modified versions of tv_grab_nz.py that get
>> their EPG files from the MythTV box web server, one for the FreeviewHD
>> source and one for the SkyNZ source.  I run the script that does this
>> from a cron job once a day at 15:10.
>>
>> I think we need to check some basics:
>>
>> 1) Is the version of tv_grab_nz-py you are running the latest 0.2.6
>> one?  If not, get the latest from here:
>>
>>   http://nice.net.nz/tv_grab_nz-py/
>>
>> 2) Have you placed the tv_grab_nz-py file in /usr/local/bin/ ?
>>
>> 3) Have you set the permissions on tv_grab_nz-py to allow execution by
>> mythfilldatabase when it is running under user/group mythtv?
>>
>> 4) What is the name of the source you are using for the Freeview
>> channels?  There should be a file in the home directory of the user
>> that runs mythfilldatabase that specifies the channels.  It is in the
>> ~/.mythtv directory and is named <your source name>.xmltv.  Because I
>> run my mythfilldatabase from a script run by user stephen, mine is
>> /home/stephen/.mythtv/FreeviewHD.xmltv, but yours may be in
>> /home/mythtv/.mythtv - I am not sure where it goes when you are
>> running tv_grab_nz-py from mythfilldatabase in the normal manner. What
>> is the contents of yours?
>>
>> 5) Have you changed the SOURCES declaration at the head of the
>> tv_grab_nz-py file to this:
>>
>> SOURCES = (
>>     'http://epg.org.nz/freeview.xml.gz',
>> )
>>
>> 6) Have you checked the spelling of the grabber specified for the
>> Freeview source.  This SQL will show you the setup of your sources:
>>
>>   select * from videosource;
>>
>> and this will show which source or sources are configured to actually
>> be used:
>>
>>   select distinct sourceid from cardinput;
>>
>> When people have been having problems configuring things and have
>> tried multiple times, there can be multiple sources configured, which
>> confuses things.
>>
>> 7) If you do this:
>>
>>   cd /var/log/mythtv
>>   grep -i mythfilldatabase *
>>
>> does it show any useful information?
>>
>1) tv_grab_nz-py  0.2.6
>2) in /usr/local/bin
>3) Permissions were owner:root group: root I have changed it to owner:root  
>group:myth (my user)
>3) I have no file with an .xmltv extension in /home/myth/.mythtv  there is  
>config.xml and tv_grab_nz-py.conf and mysql.txt
>4) SOURCES = (
>      'http://epg.org.nz/freeview.xml.gz',
>  )
>
>6) source should be called freeview   Pardon my ignorance but how do you  
>do this SQL thing?

Oops, sorry.  I have a little script that logs me into mysql and
MythTV's mythconverg database to execute commands like that.  It
requires the xmlstarlet package to be installed.  Most of the script
is just to get the right parameters for the login to mysql for all
recent versions of MythTV - the last line is the actual useful bit.  I
have attached a copy (do_mythconverg.sh).  I run it as root, but it
should also work from any user in the mythtv group.

>7) Nothing stood out in the output of that command,no errors,fails etc.
>
>Cheers
>
>Paul

You might like to try these commands to see if you can find your
*.xmltv file:

updatedb
locate *.xmltv
-------------- next part --------------
A non-text attachment was scrubbed...
Name: do_mythconverg.sh
Type: application/octet-stream
Size: 1468 bytes
Desc: not available
URL: <http://lists.ourshack.com/pipermail/mythtvnz/attachments/20130428/5e82556d/attachment.obj>