Windows Forum

Ask something
Discussion Locked

intermittent ping between 2 BDCs, W98

0 helpful
5 replies

by rpavelsf

Closed to new repliesThis discussion is closed. It stays readable, but it takes no new replies.

We have been experiencing for almost a week a very frustrating problem of 2 NT4 BDCs, one an Exchange Server, that intermittently time out talking to each other and the W98 stations on the net. We have tried everything to isolate the problem and nothing works longer than about 45 minutes.

Chronology:
Power cycled on all equipment once while testing electrical circuits.
Closet with Cayman router, Cisco PIX506, Cisco 2950 and 2924 switches may/may not have been cycled without the benefit of UPS.
4 NT Servers: jf1 PDC, Exchange and EB2 BDC all on GEB domain; EB on JFEDEB domain used only as RAS. All shut down and restarted in sequence.
New APC 2200 UPS with splitter card replaced old 1400 on JF1, EB2 and EB attached as simple connect. PowerChute 5.2 loaded on EB2.
Symptoms:
EB2 and/or Exchange stop talking to each other and workstations. Sometime a one way ping works. After anywhere between 5 and 45 minutes connections fail. After some time, connections reestablished.
Trend IISand Office on EB2 monitoring and deploying ServerProtect on all servers.

We have examined, cycled, reset Ciscos, arp, cross connected 2 servers and still had problem. Replaced a nic on one of the 2 Dell 2500 servers with no success.

All engineers are stumped. Dell says not a hardware issue. Microsoft denies software.

All suggestions appreciated.

5 replies

by estebandelatorre

Use a third w2k prfesional and launch 3 instances of pathping.exe, to both servers plus a w98 running ONLY tcp/ip. This will allows you to identify if it is a server issue or a comunication (links) issue.
Take a look to the event logs of the 2 servers and watch for any strange message.
If both servers are connected to the same switch or hub, double check if you have setting PORT MIRRORING on one port of the switch, disable it.
Monitor switch's CAM size in order to check any routing problem,if the size is near the limit, you are having routing problems!!
Monitor switch topology changes, if you have spanning tree in your structure, sometimes they take some time to get stabilized.

Just some ideas!!!

by shanghai sam

(Sorry for the delayed response; I was away on family emergency since Dec 15.)
After many days of testing and configuring the problem was resolved at the Cisco PIX. Watching packets, my consultants told me that NT was sending our some erroneous packet information that was simply relayed back by the Pix in the form of duplicate IP address for the second server. Since I did not get specific packet details, I cannot pass that information on to you. Thank you for your efforts.

This discussion is closed to new replies.