I followed this example: http://wiki.squid-cache.org/ConfigExamples/Reverse/MultipleWebservers
but can't it to work for my setup - which is a bit different but not that much. This is using a default config from Squid 2.6 on CentOS 5.4, nothing extra added or changed except for what is mentioned in the article. I get 'service is not available'...
My setup is:
test1.example.com hosted on IP1
test2.example.com hosted on IP1
test3.example.com hosted on IP2
test4.example.com hosted on IP2
test5.example.com hosted on IP3
test6.example.com hosted on IP3
test7.example.com hosted on IP4
test8.example.com hosted on IP5
testX.example.com hosted on IPX
I use Citrix Xen but I don't that's relevant as they are all separate virtual servers with separate internal IP's accessed via the Squid reverse proxy which has one external IP. The reason I am doing this is not really for web acceleration but because we don't have enough external IP's to map to each virtual server we need. A single virtual server may host more than one site, using Apache and VHOSTS. Some virtual servers may only host a single site, regardless all virtual servers will have separate internal IP addresses, naturally.
Any assistance is greatly appreciated!