OK. This may be a silly question. I have Squid set to do trans-proxy. That is going very nicely. I have, though, found that the byte-hit figures reported via SNMP are not exact as yet in 2.1PATCH2 as discussed with Glen & Henrick. On sound advice, I leave my Cisco 2503 router alone in the redirection process & use ipfwadm to do it all. This also works quite nicely. I had the thought that I might get better stats by configuring ipfwadm as follows: (1) ipfwadm -A in -i -P tcp -S 0.0.0.0/0 80 -D <my proxy IP address> (2) ipfwadm -A out -i -P tcp -S <my proxy IP address> 8080 What I thought I was doing was getting: (1) Stats for any web traffic coming into the proxy server from anywhere on port 80 (naturally!) (2) Stats for all traffic served from the proxy server port 8080 to anywhere Well, of this I'm fairly sure. However I get the results for: (1) 15 mb (2) 6550 kb This tells me that the traffic going out of my proxy is not greater than the traffic coming in by a huge factor. In fact, getting percentages at various intervals, it seems that what I am actually getting stats for is: (1) Stats for any web traffic coming into the proxy server from anywhere (2) Traffic served by the proxy on port 8080 from its cache Can anyone confirm my thoughts? Thanks in advance, Marc
