Saga of Duplicate IP's

[BobLfoot 302]

Dear Community - our plant just suffered through a dilemna that I hope never presents itself to any of you. The root cause was two Allen Bradley devices both wanting the same IP Address. The first device {and rightful owner of the address} was a CompactLogix L32E. Beginning about 2 weeks ago , it would stop communicating with the world after approximately 30 to 90 minutes on our LAN. The second device was a 9300-RADES unit which had been used as a switch in the machine of the CompactLogix before the Compact was connected direct to our LAN. Although we don't know why the RADES assummed the IP of this device connected to it and when the RADES was recycled to be used as a temporary switch and expand a leg of the LAN it comnflicted with the L32E. Just a word of caution the L32E does not have a message LED to tell you dup ID. SO you'll be searching for a while.

[Ken Roach 194]

I've run into a similar-sounding problem with the RADES in the past. As I understand it, the RADES occupies two IP addresses. The first IP address is its own LAN address, which serves as the host address for the onboard diagnostic Web server and usually as the Default Gateway address for anything attached to the local LAN. When you dial in to the 9300-RADES, your remote PC is assigned an IP address on the local LAN. Usually this is the 9300-RADES IP address +1. This is described on Page 24 of the User Manual but I think it should also be prominently noted in the Quick Start Guide. Because it is common for users to start numbering their devices at Host 2 when there's a gateway or router at Host 1, it is easy to accidentally address your controller at the same IP address as the one reserved for the dial-in device. You might not have seen the conflict for some time until a user was dialling into the RADES and the CompactLogix got power cycled. I think that the 9300-RADES doesn't use that second address until there is a user dialled into it, but I'm not certain.

[Bill Linne 0]

Interesting, Bob. Thanks for the heads-up. What I don't understand is why it took 30 to 90 minutes for this to manifest as a problem. Seems like dupe IP's would be a problem from the jump. I was expecting you to say that something was causing an on-the-fly CHANGE to an established IP, and therefore causing the conflict. Maybe someone with more Ethernet savvy can explain this for us. Bill

[BobLfoot 302]

I don't understand either Bill. The PV Plus and CompactLogix connect to one managed switch, while the RADES was connected to a second managed switch at a somewhat remote part of the plant. The two switches in question were fiber connected to the subnet gateway in the Plant Data Center. I can only assume it took a while for network data to be routed to the Compact and the Gateway to notice the dup IP's.