After multiple times of applying the changes stated in the FAQ for doing transparent proxying with Cisco and Linux using ipfwadm to redirect the cisco redirection on port 80 to 3128, I'm still getting ungodly amounts of "Connection Reset by Peer" errors. We're using squid version 2.0. This is a Dual PII 300 machine with 512mb of ram, a DPT 3334UW RAID controller with a 4gig seagate barracuda system disk and 3 Seagate Barracuda 9 gig drives in a RAID 0 configuration. Kernel 2.0.36 with all options configured correctly for the makefile Here's the squid configuration : > # Squid.conf - Squid Proxy Server config file > # BK 11/4/98 > > # Port Response > http_port 3128 > httpd_accel_host virtual > httpd_accel_port 80 > httpd_accel_with_proxy on > httpd_accel_uses_host_header on > > # Cached Memory Size > cache_mem 256 MB > > # Cache Directories > cache_dir /var/spool/squid 18000 16 256 > > # Log Directories > cache_access_log /var/log/squid/access.log > cache_log /var/log/squid/cache.log > cache_store_log /var/log/squid/store.log > > # Running Squid Pid file > pid_filename /var/run/squid.pid > > # Squid Options > cachemgr_passwd ********* config > cache_mgr root > redirect_children 15 > dns_children 10 > authenticate_children 15 > announce_period 1 day > announce_host sd.cache.nlanr.net > announce_port 3131 > reference_age 2 weeks > request_timeout 60 seconds > memory_pools off > > # Access Control List > # BEGIN > > acl manager proto cache_object > acl ns2 src 209.172.0.8/255.255.255.255 > acl all src 0.0.0.0/0.0.0.0 > http_access deny manager !ns2 > http_access allow all > > # Access Control List > # END > > # Cache stops > #hierarchy_stoplist *.asp > > # Cache > # > # Format is: hostname type http_port udp_port > # > cache_peer sv.cache.nlanr.net sibling 3128 3130 > cache_peer sj.cache.nlanr.net sibling 3128 3130 Any help you can offer would be much appreciated. Ben Kohn Manager of Information Systems Triton Technologies Inc.
