Cisco Support Community
Showing results for 
Search instead for 
Did you mean: 
Community Member

FastEtherchannel, STP problem with Cluster servers

We have 2 servers that are clustered with 2 NICs each for the network and 1 dedicated NIC each for replication. The 2 NIC connections for each server are configured for FastEtherchannel on the cards and are plugged into separate switches for redundancy. When looking at the diagnostic logs on the 4 switches involved (Catalyst 3500's) each port that the servers are connected to display the following message %RTD-1-ADDR_FLAP: FastEthernet Mod/Port Relearning x addresses per min. These messages are filling the logs and it now seems that we are experiencing some network slowdown. Should spanning-tree portfast be disabled on the 4 ports involved? No port groups were setup on the switches ( I am not presently sure if it is possible to setup port groups between ports on different switches)

Cisco Employee

Re: FastEtherchannel, STP problem with Cluster servers

If the 2 NICs connect to two different switches, configuring them as a channel on the NIC could be the issue. Try disabling NIC card teaming and see if it makes any difference.

CreatePlease to create content