cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
249
Views
0
Helpful
1
Replies

C3512XL ADDR_FLAP messages

Danilo Dy
VIP Alumni
VIP Alumni

I have two C3512XL Switch, 2 NOKIA FW, and 2 PIX FW.

PIX is connected to FA0/1

NOKIA is connected to FA0/2

FA0/12 is Trunk

The network become very slow and I received this messages from two switches;

%RTD-1-ADDR_FLAP: FastEthernet0/2 relearning 14 addrs per min

When I execute "debug ethernet-controller address" here is the messages;

*Jun 9 00:11:41: Add address 0000.5e00.0123, on port 12 vlan 1

*Jun 9 00:11:41: 0000.5e00.0123 has moved from port 1 to port 12 in vlan 1

*Jun 9 00:11:42: Add address 0000.5e00.0123, on port 1 vlan 1

*Jun 9 00:11:42: 0000.5e00.0123 has moved from port 12 to port 1 in vlan 1

*Jun 9 00:11:46: Add address 0000.5e00.0123, on port 12 vlan 1

0000.5e00.0123 is a MAC Address of NOKIA VIP using VRRP. We don't see any problem with NOKIA.

Cisco has this recommendation;

http://www.cisco.com/en/US/products/hw/switches/ps628/products_configuration_guide_chapter09186a008007e8ec.html

Thus it means to disable port FA0/1? But we need that PIX FW. Any other solution for this?

Thanks in advance.

1 Reply 1

Danilo Dy
VIP Alumni
VIP Alumni

Okay, I solved the problem. I shutdown PIX 2 and NOKIA 2 (those that are connected to switch 2, which all the traffic passes because PIX 2 and NOKIA 1 is active, whereas PIX 1 and NOKIA 2 is standby). The error messages is gone but the high latency is still there.

I replace switch 2 with a new one. Power up all FW, all the error messages is gone and now the network is in optimum performance.

I send the switch for RMA. This is the third 3500 series switch that fails after two years of deployment (we bought it brand new 2 years ago), the switches fails whithin 1 month of each other (like a domino). It seems, switching is more difficult than routing :) the switches fail more often. The funny thing is that the first switch that fails, the new switch that we temporary put in until the RMA arrive also fails within 1 week.

QUESTION: If i have a brand new router and it fails and I send it for RMA. Should Cisco replace it with a new one or an old working one?