Hi there,
I'm in the process of designing for a stretch cluster across two sites with two independent fibre links between the sites and trying to figure out the implementation details required for the MLAG configuration. This is easy with stackable switches on each site as there's just the one LAG to create, but here I think I need to create multiple ones. It's a closed storage network so the LAGs can pass all VLANs so that I don't need to touch them if I add VLANs to the other ports later on.
One of the things I'm fuzzy on is whether the peer-port configuration actually passes traffic or whether it's just for state coordination. I.E. Do I need to use sfp2 for the peer-ports in addition to the existing interswitch LAG or can I consol
The servers are ESXi so I'll be using the built-in load balancing and anything else I'll just use active/passive aggregates so I won't need to do any LACP/LAGs between the two switches on a given site.