[j-nsp] Virtual Chassis Fabric ports and LAG

[Jonathan Call]

I've set up four QFX devices in a lab. Two QFX5100-24Q as the spine members and two QFX-5100-48T as the leaf nodes. I connected two 40GbE links from each spine to each leaf member. According to the documentation the VCPs are supposed to figure out a LAG on their own. I can't tell for certain if this is happening.


In a standard Virtual Chassis among EX switches when VCP ports become bundled they are assigned a positive integer Trunk ID. The Trunk IDs in the "show virtual-chassis vc-port" on this VCF cluster are all -1.  Besides pushing a ton of data through the links is there some way to confirm if LAG is actually happening?


Jonathan