05-08-2012 09:25 PM
I'm looking for information on how traffic flows out of a VCS fabric towards a MAC at the end of a vLAG.
I understand well that from the perspective of Host X, it's using plain old LACP to bring up the LAG. From its point of view A and B are actually one and the same switch. My questions are about VDX-C and what it sees. What does C have in its forwarding table as the next-hop for X's MAC address? Does it have equal cost routes to reach X? With the vendor implementation I'm familiar with, A and B create a pseudo-node that logically sits in-line between X/A and X/B. A and B then advertise this pseudo-node into the fabric as the Rbridge for reaching X. When C then looks up X, it finds that it needs to send frames to the pseudo-node. When C recursively looks up X's Rbridge ID, it finds two equal cost routes (A and B). Does VCS operate in a similar fashion?