13w6d: %RTD-1-ADDR_FLAP: GigabitEthernet0/1 relearning 5 addrs per min

[a.mir]

the problem started couple days ago, some of the switches down, i am thinking its pointing towards a loop in the network.Please advise what could be the actual cause and the best way to troubleshoot this..Thankss

[tdrais]

Not sure if you can read this link since it is in TAC

http://www.ciscotaccc.com/kaidara-advisor/lanswitching/showcase?case=K19174025

All it means is that a mac address is moving back and forth between ports quickly. It is not always a problem if for example if you have wireless access points but in your case I would bet it is a spanning tree issue.

[royalblues]

Its true that it may be caused because of a loop.

The best way to check whether it ia a spanning tree loop is to check the mac-address-able aging time. If you see that the aging time is 15secs most of the time instead of 300secs, then its a possibility.

These errors are also produced by servers who are teamed together or running NLB.

This can be seen by issuing sh mac-address-table interface command wherein you get multiple MAc address entries

Narayan.

[a.mir]

Thats correct, the loop with multiple entries after a long finding, there was a netgear switch with both ends of the cable in the same switch caused a caos.Finally able to find it. Thanks a lot guys.