Hello,

Recently, two of my Novell servers (in the same subnet) began displaying ip conflict errors. The MAC addresses listed are the addresses for their own NIC's.

We had brought a new server in a different subnet but in the same Tree online a couple of days before this started - no other network or server config. changes.

Help....?

Thanks,
Tony Baldwin

Look at your DNS records.

Hi,

With some smart switches, the IP and MAC will be recorded for a specific port. Altering the RJ45 connection port may cause the switch to scan itself and try to compensate to resolve issues. Its stored IP/MAC can then conflict with the true IP/MAC and a server will have a IP conflict with itself.

The best fix for this would be to reset the switch (per manufactuer's instructions) to relieve any stored IP/MAC combinations, and eliminate the conflict that never existed in the first place.

I don't know if this is what you're experiencing, but I'm passing along the information regardless.

HTH,
-J

Thanks for the ideas, based on your suggestion, File Clerk, I did reset the Cisco switch (even put an unmanaged switch in its place temporarily to test) but it did not resolve the condition.

As to what the wanderer suggested, how do I do that on a Novell 5 server? I am fairly experienced at this, but am not sure wht you meanor how to do it.

Thanks agian,
Tony Baldwin

Hi Tony,

I wasn't very descriptive before, as I had no idea if you were monitoring it. As from my own experience, I really think the switch or hub is still to blame.

The following scribbles are step by step fixes for me in the past:

Fix #1:

1. Unplug the RJ45 from the server.
2. Flush the routing tables, just because.
3. Change the switch or hub duplex. (If full set it to half - if at half, set it to full. As you have said you have Cisco, I don't believe a single toggle switch is the case, but it might be. If both are set equal (usually to HALF) it should work.)
4. Reconnect server and it should be fine.

Fix #2:

1. Disconnect server from the switch.
2. Rebind the IP.
3. Flush Server's routing tables.
4. Disable ARP on the switch.
5. Reconnect and it should be fine.

I strongly feel one of those two will get you back in action, if not, post again.

Nothing is fixed so good that it cannot be fixed more! ;)

HTH,
-J