Peer Cache

[lbb87]

I've been wondering about something for a while now.

How exactly does the peer cache work? I read that there are/were three different peer cache servers running, all using different host files/patches. I've seen each of the different host files and since they all use the same IP addresses, I don't understand how you could have three different peer cache servers.
One host file had c3310.z1301.winmx.com with an ip address of 205.238.40.1 and the other had c3310.z1301.winmx.com with an ip address of 82.195.155.5.  What in the heck is c3310.z1301.winmx.com - what does that do?
I know that the host file method isn't the most popular choice right now but I refuse to use a newer method until I understand more about what's going on. I hate using something and not having at least a general idea of how it works.

So... Could someone please explain the peer caches to me?


I'm starting to wonder if maybe I should switch host files. Maybe I would get better search results as I'm not really getting as many results as I was a few weeks ago. I'm on dialup so I'm obviously connecting as a secondary. I'm also using PG2 and have all the proper blocklists in place. Of course, maybe the problem is that I'm searching for rare files. They sure are hard to get a hold of.

[Bearded Blunder]

c3310.z1301.winmx.com is the hostname (friendly address) of one of the original winmx peer cache servers.

Before they shut down what happened was winmx used the same system to look this up to find it's IP address that your internet explorer does (DNS)

The hosts file fix works because windows has a file it looks in for this information BEFORE it looks online to find the IP.. called HOSTS, it's really just a list, as servers come & go this list needs updating.. There are consequently a number of versions of this around, with caches added (more ip's) etc.

The Winsock patch ws2_32.dll works differently, instead of a fixed list each user has to keep up to date it translates something like c3310.z1301.winmx.com to cache0.winmxworld.com  this then gets looked up using DNS, which can be always kept current, without you doing anything manually, so if you're using the .dll you can be sure you're not trying to use any cache that closed or moved IP.. or that if there's a new one, your system knows about it (because DNS does).