Hi all,
Av gone through many of the posts here relating to squid and I don't seem to get the solution I want well.

The server (Fedora Core 6)am running squid on is also my DNS server so I don't expect high delays when any of the clients are querying for dns.
Client computers on this setup are either going through the proxy server directly, transparently, while others are directed through the web browsers.

The squid version am running is:squid-2.6.STABLE6

Here is an extract of the Config.s:
####

http_port 3128 transparent

acl LAN_A src 192.168.20.0/24
http_access allow LAN_A
acl LAN_B src 192.168.30.0/24
http_access allow LAN_B

visible_hostname "MYHOSTNAME"
cache_dir ufs /var/spool/squid 2000 16 256

Having squid as my cache server with a 2gig cache, I expect to get pages av already visited through my LAN instantly the next time I visit without much delay.(Something close to that exhibited by web-browsers when they are caching pages) but the delays last for longer period than I expect.
How best should I configure squid to have robust response speed during requests or then;
Is there any other caching server/software that can meet my desired situation perfectly?


Please advice.

You are using transparent proxy. Do you also have the correct iptables settings for this. You need to narrow down where your problem is coming from.

All the redirection to squid in iptables for all the LAN using squid transparently is ok.And the clients go thro' squid successfully - the squid access log confirms this perfectly.
The issue I have is - Is there any way to configure squid to have a high response time?