We have an IPCop box which has been on our network for a number of years now and generally works well. Up until recently, it was only accessed by a limited number of machines, but now we're setting it as the gateway for everything on the network.

The problem is that it doesn't seem to respond to certain internal IP addresses. It drops all requests from 192.168.1.26, for example and won't even respond to pings from that IP, but will route requests from 192.168.1.24 with no trouble.

I can't see anything in the firewall rules on the IPCop interface which looks relevant and I've also tried iptables. The only entries I can see relating to private IPs are ones which should be there and don't relate to the IPs in question.

I have a feeling that this was set up intentionally by a former network admin a number of years ago, but I've no idea how he's done it.

Neither networking nor linux are among my specialities, so I'm not really sure what to try next.

Just had a notification from the forum to suggest that I update with additional information or subsequent steps I've taken, as the thread has not had any replies after five days.

I haven't really taken any further troubleshooting steps, as I only posted this once I'd run out of steps to try, but I do have another question which might take me in the right direction if anybody's able to answer it: Are there any gotchas with iptables that might be catching me out? Or any other parts of Linux/IPCop I don't know about, which might be doing some kind of filtering?

I've worked on the assumption that these packets are being dropped by the IPCop box itself, because machines with IPs that it doesn't like are still able to communicate with other devices on the network, but is it possible that this is happening because of the config of something else, somewhere else on the network?