E1 OpenPLi SQLite XML EPG Plugin (Radio Times)

[PaphosAL]

Here you go, Bob- special edition just for your setup.

Note that the comments I made in point 4. of the upgrade readme apply equally to your setup. It's not a good idea to keep appending to the existing 'fat' epg.db on your Win mount at 06:30 daily. This results in a dbase file larger than necessary, and, nore importantly, takes about 50% longer than needed.

So make sure you have /var/etc/epg.base present (from the upgrade package), so that the EPG load starts afresh. That file is similar to the one you saw yesterday in the /tmp folder. Early on in the script, the existing (large) epp.db gets overwritten with the small 'starter' file, before launching the download / filter/ convert loop.

It basically replicates a standard OpenPLi setup, where a user carries out the recommended reboot prior to Retrieval of the new EPG. So the large /tmp/epg.db vanishes during the reboot, and the new small one automatically appears afterwards. On a mount setup, the large file remains, so we must force things using the overwrite method.

Please post tomorrow morning's log, so I can compare before and after results.

Cheers- AL

[bobsylas]

Thanks for this Al
copied file over & set cron reboot for 05:30
will post the log file tomorrow

Cheers Bob

Bob- please report on the epg.db file size as well, m8. AL

[bobsylas]

Hi Al the epg.db file on my windows machine is 7.386 KB
all is working great

here is the log file

Cheers Bob

[PaphosAL]

That's a good result, Bob, with the whole process taking over 2 minutes LESS to complete than before.

Tell you what- while Auntie is being so reliable on a day to day basis now (last Sunday's 24 hour outage being the first blip since before Christmas), you might as well leave your setup like it is for the time being, downloading direct from the RT site.

But if RT becomes 50/50 again, then we'll look at moving you onto the method I use on my server.

Cheers- AL

[bobsylas]

Hi Al don't know if it's my set up or not but my EPG stopped working a week or two ago ?

my epg.db file now is only 79 kb as opposed to nearly 9 meg

the only way I can get it to work now is by using the original rt_uk.sh file but can't see what has changed

cheers Bob

[PaphosAL]

Logfile please Bob? All is fine here, but I have a sneaking suspicion what is causing your issue.

Cheers- AL

[bobsylas]

A couple of log files from yesterday Al

Cheers

[PaphosAL]

Hmmm, this log suggests to me that at 9pm last night, all dat files on Auntie's server were 0 bytes, ie, a technical fault at their end.

Can you repeat this test, first with LraiZer's new script? If this fails, save the log, then try again with the old script. If that succeeds, let us have a rar of both logs AND the new script, so I can compare it with my script here, please.

Also, try running your cron job a bit later in the morning, at a time when nobody is ever watching telly. I suggest this because RT is updated on their server at 06:45 GMT, which is of course 07:45 at the moment.

So it's possible that your 07:00 cron might be catching Auntie with her knickers down, ie: at a maintenace point where all the old dat files are emptied, in readiness for the new 14 Day lot.

Cheers- AL

------------------------

Edit:

Hang on a mo, Bob- here is the command line in LraiZer's new script that actually downloads each of your mapped RT dat files:

 
process=`time -p wget -q "$url/$chanId.dat" -O "/tmp/$chanId.dat" 2>&1|grep real|sed 's/real.//'`

time is a binary (25kb chmod 755) that should still be in your /var/bin folder. Without it, I would imagine that the whole command line will fail. Have you deleted it, forgetting what it's for, maybe? If so, grab it out of the modded kit and try again!

[bobsylas]

Originally Posted by PaphosAL

time is a binary (25kb chmod 755) that should still be in your /var/bin folder. Without it, I would imagine that the whole command line will fail. Have you deleted it, forgetting what it's for, maybe? If so, grab it out of the modded kit and try again!

wow looks like your right Al, did an image update a couple of weeks ago & didn't copy the "time" file over
still took about 13 mins so here's the log

[PaphosAL]

That's taking it's time, Bob! No rational explanation for the slowness in the log, either- all the download times look fine to me...

My bet is that you're appending new data to the old 'fat' /tmp/epg.db file, because you forgot to copy epg.base (79kb) into /var/etc