[squid-users] accelerator and multiple web servers

I'm attempting to configure squid to accelerate my intranet web site to make it available to remote users. I've managed to get the basic setup working properly - I can get at the main intranet server (call it intra) without trouble. My setup looks like:

Client (using URL http://externalhost.twr.com) ---> (a.b.c.d) firewall (x.y.z.a) -----> squid server (x.y.z.b) ----> intra (x.y.z.c)

The problem begins when the intranet site references other internal sites, say "bugs" for bug tracking. The URL that the client gets back contains a hostname that does not exist in the public DNS (and wouldn't be directly available even if it was).

So, what I'm trying to do is get squid configured to accelerate multiple web servers on multiple machines:

Client (using URL http://externalhost.twr.com) ---> (a.b.c.d) firewall (x.y.z.a) -----> squid server (x.y.z.b) ----> intra (x.y.z.c)
                                                                                                                               |
                                                                                                                                -------------------------> bugs (x.y.z.d)

All these servers are on separate machines.

I've read what I could find in the various FAQ, User Guide, mailing list archives, etc. but I'm a bit confused.

I'm not sure if what I need to do is use a redirector or virtual hosts or httpd_accel_uses_host_header? Do I need to have the external DNS point "bugs" to the externalhost IP?

Anyone currently doing this who can share some wisdom?

Thanks,

Stephen
Received on Thu Apr 12 2001 - 08:54:25 MDT