Journal

My own little TV Station

· Geek

We have been working on a pretty neat little project, here in Edtown.

Using a pretty high powered desktop machine, we have been working on a custom setup to stream Educational “Free to everyone” TV stations over our network. Right now to stream 4 channels it looks like it is going to take a P4 3.4Ghz 2mb Cache. That is transcoding and streaming them all, all the time.

On the other end, to decode one channel at a time, it looks like it is going to take a PII 233mhz. Both machines will be running Linux and running VideoLan-Client to do this. VLC is pretty cool. You can do a whole lotta stuff with it. There are apparently colleges that are streaming DVDs across their network and stuff with it.

I also setup a yam server. No maxx not just because I wanted to be like you. Since we run a proxy server and all our machines have to route through it, it was making it a real PITA to get the packages for my machines. YUM really doesn’t seem to like going through authenticating proxies so I just decided to cut out that step and move all the packages in house.

I am having a really odd error on that though were some of my repositories keep looking in “RPMS.$repo/Fedora/RPMS/” instead of just looking in “RPMS.$repo” for the packages. maxx you having any of that?

I got some howtos up about setting the VideoLan Streaming machine and a howto on how to setup the frontend clients too

2 comments

Comments are closed; these are preserved from the original blog.

  1. maxxJuly 22, 2005

    No Maxx, not -just- because I wanted to be like you So... that would mean that -part- of the reason you set it up is because you wanted to be like me! :-)

    No, I don't have that issue with Yam.

    And it looks like I'll be setting up the VideoLAN thing as well. May have some Q's for you next week.

  2. maxxJuly 23, 2005

    So I found the problem... I had a cron job that I setup -before- I realized yam had an init script - and the script had been running every night and not completing. So there were about a dozen 'yam -uxgq' jobs in the queue (at 50kB/s each) which would cause the pipe to become saturated every time the -legitimate- yam job started up. A purge of the cron job and all the rsync processes fixed it. So it wasn't rsync or yam that were to blame.