We have one primary and one redundant core 6509. The core 6509’s are link together via the GBIC port on the supervisor cards. Each of our access layer 3500’s have one link to each of the 6509’s. HSRP is running between the core 6509’s. Spanning tree is running throughout the network. Both of the core 6509’s have the same 24 VLANs defined on their ports. All ports on the 3500s are members of whatever VLAN the link to the core 6509 is in. I am not able to view the configuration of the 6509’s. I assume we are doing multilayer switching however.
I am wondering how traffic gets routed or switched between the core 6509’s in the following scenario:
The redundant links on two 3500’s are in blocking mode. Each 3500 is in a separate VLAN. The primary link on 3500 goes down causing it’s redundant link to go into forwarding mode. Another 3500 has a forwarding primary link and a blocked redundant link. Hosts patched into the two 3500’s now need to communicate.
I don’t see HSRP entering the picture since in this scenario there has not been a 6509 failure. I thought perhaps Multilayer switching was taking over. However, I read that MLS with multiple MLS-SEs and MLS-RPs in separate switch chassis is not supported.
