We have 6 catalyst 4500 switches working as routers where one of them is considered as the active HSRP router and the root bridge for spanning tree. We faced slowliness in the network last week for half an hour where things get back to normal after that. After checking the switches logs, we foud that all the six switches were giving host flapping error messages for multiple MAC addresses. The error log has the following format:

Jun 22 00:17:49: %C4K_EBM-4-HOSTFLAPPING: Host xx in vlan 50 is flapping between port Fa2/47 and port Gi3/1

where xx is the MAC address for different hosts.

The following error message was observed in the primary and secondary HSRP routers only:

Jun 22 00:18:18: %HSRP-4-DUPADDR: Duplicate address xx on Vlan4, sourced by yy

where xx and yy are ip and mac addresses respictively.

It seems that there was a looping issue or HSRP flapping that caused these error messages and flooding to the network. However what caused these problems to occur although the network was stable before the problem and get back to normal by itself after half an hour.

My conclusion is that there were some ports going up and down which causes continous spanning tree recalculation and HSRP flapping. Is that conclusion right? And if it is right what causes some ports to go down and up repeatitly for certain time. Did you face anything like this before. Is this GBIC problem? switch port problem? cable problem?