Power Monitor 3000 communication over ethernet

[msynco 0]

I could sure use some help, if anyone can point me in the right direction, I'd appreciate it. Our plant has a network of 28 AB pm3000's with ethernet comms. Approx. 2 weeks ago, all but 8 of them quit communicating. A few facts about the system: 1. Has been in service for 8-10 years 2. Power monitors share the same fiber optic and switches as our plant's network, but are on a different network. 3. There are 28 AB SLC505's on the same network 4. When powered down, then back on, the 20 that dont work, will go online for about 5 minutes, then drop back off. 5. The meters themselves are still gathering data, just the RSLinx connectivity goes away. 6. When looking at the RSLinx diagnostics , I see a lot of communication timeouts. 7. IT department that maintains the network has made no changes to the switches(routers) in the immediate past. 8. I've always had the occasional meter that would lose connection, and have to be power cycled to restart comms. 9. The 8 meters that do work are using the old firmware in the meter, Base FRN 2.44 and Ethernet FRN 1.21, while the ones I've checked that dont work are newer firmware.(havent checked them all, but it does appear to be a pattern relating to firmware) 10. I took one of the working meters and upgrade to the newest firmware from the AB website. It now will not maintain communications more than 5 minutes without cycling power. I've got the Plant manager breathing down my neck to get the numbers coming back in, and have run out of ideas to check. Any ideas or suggestions will be greatly appreciated. Thanks, Mike

[TimWilborne 108]

Sounds like a timeout difference between the old and new firmware but that's strictly a guess based off of what you said. Ken or one of the other Ethernet gurus would probably be better ones to help on this one but a few basics that you may or may not have covered. When you ping the devices in question from the RsLinx PC, do their delays seem any different than the working devices? Are you seeing any dropped packets? Any crazy hops on tracert to the IP? Probably more of a web problem but I guess it could crop up on a large Ethernet network if IT was shuffling things around. After that you are probably into Wireshark territory and I only know enough to collect data and send it to someone else to help troubleshoot.

[msynco 0]

When the PM's are power cycled the ping returns a normal result until they drop comms. Our IT department has been monitoring the network, and reports they can find no problems. On a side note, when the trouble first started, we had a an ethernet I/O module that was flooding the network with multicasts so much that the router was unable to maintain it's routing table. I have removed power to the device, and will leave it off until we resolve the problem. Mike