[squid-users] TCP_MISS/403

Hello,

sorry for this post, i know, there are more mail in list
archive, but i didn't find for me the answer.

sorry for the long mail.

i have a FreeBSD 4.4, Squid 2.5 PRE5, in our Intranet.
Squid ip's is 10.0.100.251.

Here are two networks, what squid works for: 10.1.0.0/16 and
10.0.0.0/16, but second network fragmented more little networks.
(forexample: 10.0.4.0/24, 10.0.5.0/24...)

every ip connection works correctly, from everywhere. (icmp,
ssh, etc...)

HTTP connect works correctly from 10.1.0.0/16, but does not
works for any hosts, here is sample log:

1016920321.721 1216 10.0.4.60 TCP_MISS/403 ... http://www....

in squid.conf _is_not_ any deny acl. (nothing deny, all acl
are allow!)

the different between two networks (example: 10.0.4.0 &
10.1.0.0) is routing, but here are two traceroute output:

proxy:~# traceroute 10.1.1.30
traceroute to 10.1.1.30 (10.1.1.30), 30 hops max, 38 byte packets
1 10.0.100.249 (10.0.100.249) 1.235 ms 2.031 ms 1.167 ms
2 10.0.100.254 (10.0.100.254) 8.940 ms 6.652 ms 2.234 ms
3 10.0.100.10 (10.0.100.10) 18.052 ms 10.0.100.5 (10.0.100.5) 47.025 ms 40.766 ms
4 10.1.1.30 (10.1.1.30) 48.802 ms 32.628 ms 48.742 ms
(this works correctly)

proxy:~# traceroute 10.0.4.60
traceroute to 10.0.4.60 (10.0.4.60), 30 hops max, 38 byte packets
1 10.0.100.249 (10.0.100.249) 0.859 ms 0.816 ms 0.966 ms
2 10.0.4.60 (10.0.4.60) 0.610 ms 0.655 ms 0.487 ms

or:

proxy:~# traceroute 10.0.13.11
traceroute to 10.0.13.11 (10.0.13.11), 30 hops max, 38 byte packets
1 10.0.100.249 (10.0.100.249) 0.820 ms 1.850 ms 1.617 ms
2 10.0.100.41 (10.0.100.41) 16.440 ms 16.250 ms 16.223 ms
3 10.0.13.11 (10.0.13.11) 19.553 ms 17.611 ms 17.794 ms

in first case is a redirect, there is an other router.

what is the problem?

please help me, it is very hard problem.

thank you:

a.
Received on Sat Mar 23 2002 - 15:06:31 MST