Hi, I'm trying to help design a proxy setup for a client. I'm not too familiar with Squid, they are, but they aren't familiar with VPNs and Cisco routers, that's my job. I've looked through the archives and haven't found a good working example of what we are trying to accomplish. Here is the setup:
We have a central office with 2 Squid 2.5 servers / FreeBSD 4.4. We have over 100 branches that all have their own Internet connections through various means. All of the branches are connected to the central office via VPNs to a Cisco 3030 VPN Concentrator allowing them access to the centralized servers and other network resources. The central office has a 15 meg pipe to the Internet, most of the branches are using ADSL with 384k upstream. I've got questions, maybe somebody out there can help direct me in the right direction. With this scenario, is it possible to use the two Squid servers in the central office to service all of the clients across this VPN network? (We plan to divide the clients evenly to the 2 Squid servers). HTTP traffic outside of the network is not considered to be terribly important, so even though it will be quite slow, will it be TOO slow? Should we be using WCCP? (We want to proxy transparently.) Are there other alternatives to this? Should we even be considering doing this? Thanks, any advice is greatly appreciated! Michael
