See Who's Causing ADDR_FLAP Messages
Posted by Tres Mon, 03 Jul 2006 14:45:00 GMT
Normally on Cisco switches, MAC addresses are learned once per a port; however, if the same MAC address is seen on two different ports, it will cause an ADDR_FLAP error to be printed to the log. This normally happens one of two ways. If there is a port anywhere in the switched intranet that’s looped back to itself, addresses will jump between the real port and the port that’s in the path to the looped port. Or, if there is a problem with a redundant PF CARP firewall which causes both routers to be in MASTER mode for a given CARP port.
When this happens, the error reads:
ADDR_FLAP, RTD, LOG_ALERT, 0, "%s relearning %d addrs per minWhich is nice to know, but while Cisco Switches may tell you that there’s someone causing problems, it won’t tell you who it is unless you turn on debugging output.
In order to get the MAC address of the host causing the problems, do a:
debug ethernet-controller addr
And when you’re finished, you’ll need a:
no debug ethernet-controller addr
to turn debugging off.