%RTD-1-ADDR_FLAP: GigabitEthernet0/2 relearning 5 addrs per min

[ciscosheryl]

Receiving this error message in 4 different switches in the same switch closet. They are not clustered together:%RTD-1-ADDR_FLAP: GigabitEthernet0/2 relearning 5 addrs per min. Debug ethernet-controllers address shows many additions and deletions on the trunk ports going back to the core. I also show the same mac address learned on different gig interfaces (say gig 0/1 and 0/2) and on different vlans (see attached). Any ideas on next steps to resolution?

[Prashanth Krishnappa]

Most common reason is a STP loop. Check for any potential layer 2 loops if you have redundant connections.

[scottmac]

Got any wireless or VPNs that could be an unmanaged loop?

Make sure Spanning-tree is up everywhere.

Start looking around for some user with a switch/AP under their desk.

Maybe one of your switches lost their trunking configs .... so now segments that used to be different VLANs are actually dumped into one big happy VLAN (i.e., have you added a new (Cisco) switch to the network recently?

The old "adding a new (unconfigured / non-VTP-configured) switch and having the latest VTP serial number push the (non-configuration) config out to all of the other switches on the net" thing come to mind (all/some of your VLANs went away).

Good Luck

Scott

[ciscosheryl]

Thanks Scott. Yes, we have added some 3750's lately, but, not in this location. I'm going to look into the "switch lost one of their trunk configs." thanks again.