I have a box (gateway) between two subnets, 192.168.x.0 and 10.0.0.0 I can ping from gateway to 10.0.0.138 and connect to the webserver on 10.0.0.138 from the gateway (lynx).
I can ping from the 192 subnet to 10.0.0.1, but I cannot ping or connect to 10.0.0.138 from 192.168.x.0 subnet. ip forwarding is on on gateway and there are presently no firewall rules in place. gateway's routing table is like this Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 192.168.1.0 * 255.255.255.0 U 0 0 0 eth1 10.0.0.0 * 255.0.0.0 U 0 0 0 eth0 127.0.0.0 * 255.0.0.0 U 0 0 0 lo default 192.168.1.254 0.0.0.0 UG 0 0 0 eth1 tcpdump -i eth0 on gateway (while another machine is trying to ping from 192.168 etc) is like this: [root@gateway ipv4]# tcpdump -i eth0 Kernel filter, protocol ALL, datagram packet socket tcpdump: listening on eth0 13:03:34.988970 > arp who-has 10.0.0.138 tell 10.0.0.1 (0:0:c0:56:70:b0) 13:03:34.988970 < arp reply 10.0.0.138 is-at 0:90:d0:6:a1:d5 (0:0:c0:56:70:b0) 13:03:35.178970 > 192.168.1.23 > 10.0.0.138: icmp: echo request (DF) 13:03:36.178970 > 192.168.1.23 > 10.0.0.138: icmp: echo request (DF) 13:03:37.178970 > 192.168.1.23 > 10.0.0.138: icmp: echo request (DF) 13:03:38.178970 > 192.168.1.23 > 10.0.0.138: icmp: echo request (DF) ie never any replies (except to the arp). Connections to port 80 of 10.0.0.138 are also rebufffed. Whats the guts? (gateway is running linux of course. The 10.0.0.138 is an adsl modem. -- Nick Rout <[EMAIL PROTECTED]>
