[skhosla]

I was wondering if it is possible to have a completely seperate box build the database, run the scrapers and download all the fan-art.

My setup is like this. I have a central server that has a huge amount of content, and lots of free cpu and ram. It runs gentoo.

Code:

skhosla@neutron /array/archive/TV_Shows $ find . | grep -i mkv | wc -l
1952
skhosla@neutron /array/archive/TV_Shows $ find . | grep -i avi | wc -l
37030
skhosla@neutron /array/archive/TV_Shows $ cd ../Movies
skhosla@neutron /array/archive/Movies $ find . | grep -i avi | wc -l
5104

I would then eventually have about 30 xbmc boxes connect to this server to get the content. The xbmc boxes will have less processing power/ram etc.

What I would like to as a first step as have the server index all the existing content, scrape all the additional info, get the artwork etc, so that when a new xbmc box is made, the data can just be copied over to it.

As a second step, if it is possible, I would like to see if the central server can add the info for new content and push that out dynamically to the xbmc boxes (ie. rsyncing the .xbmc directory or trying direct writes to the sqlite database - will require some experimentation).

So is there a way on the central server I can index all the existing content and get it ready to push out to the client boxes?