Hi,
I'm going to try to explain this the best I can to see if anyone has had a similar issue or knowledge in this subject. My workstation is running opensuse 10.3 and it sits in our main office. We have a colo site that is connected to our main office with a 50 meg fiber connection. The colo is connected within the same subnet (we don't route to it but bridge it). For example my workstation is 192.168.201.6 and a colo server is 192.168.203.50 and the subnet mask is 255.255.252.0, so it's basically a supernetted class c. In order for a colo server to communicate with the main office, it goes through 4 switches, 2 at the colo and 2 at the main office. This works fantastically with everything at the colo and the main office...except for my linux workstation. I can't ping or telnet or ssh anything at the colo, but I can hit everything here at the main office. I have my subnet mask right, and I can ping networks beyond the colo (we have some wans connected beyond the colo). Something else interesting is that when I try to ping that server (192.168.203.50) I get the right entry into my arp table and when I look at the server at the colo it has the right arp entry in it's arp table as well. And when I look at each of the switches in between the server and my workstation they all have the correct mac address table entries for my workstation. I'm completely confused! Is there something huge I'm overlooking?

Thanks,

I almost forgot to mention that I'm running a windows xp vm on my workstation that is using the same network card. I can hit everything at the colo from that vm.

What happens if you turn off your firewall and run these tests?

Forrest

Thanks for the reply. I figured it out...it was a rogue management interface with a mis-configured subnet mask on a cisco router that was answering to arp requests. Now communications to the dr site works from my workstation and much faster from the others.