To those who care/interested,
I was looking at Napster the other night, I wanna be able to do some non-regular stuff, IE: find the biggest mp3, find more than 100+ matches, search by file size etc..
Packet sniffing a enthernet connection running napster shows the searches and finds are via SQL statements.
I'm not exactly sure how napster downloads files at this point.
Here's what I know so far about Napster and its protocol...
1). Client opens and Napster beings to connect
2). Napster connects to a database/webserver and gets an address of a load balancing server.
3). Napster connects to that destination machine and the machine (load balancing server) spits out a random IP address and port for the Napster client to connect to.
4). Napster connects to that random IP/port and uses IRC as it's protocol/transport.
From there on, searchs, list updates, chats, requests etc are done via standard IRC communcation. (Check out some RFC's if you wanna know what this is)
After playing around I can get a random IP from the load balancing server and its port, then connect to it through IRC, however the server is silent. Meaning it wont say something like "Hello, this is ChatHost 21.2xx running on blahblah.webmonkey.com for IrcD.Linux242.openHost." something to that effect. The server as soon as you connect to it is waiting for some sort of command/login text to be sent, but I can't figure out what. At a guess, the files are also transferred via DCC (IRC File Transfer Protocol).
Anyone wanna help me pull it apart for something to do?
Any help at all would be appreciated.
