Nexus L3 link over VPC trunk but in separate non VPC SVI/VLAN?

[wintermute000]

Trying to clarify this statement


To build L3 backup routing path, use the following options listed by descending order of preference:
● Use a dedicated Layer 3 point-to-point link between the vPC peer devices to establish a Layer 3 backup path to the core.
● Use the already existing Layer 2 port-channel trunk ISL (Inter Switch Link) for non-vPC VLAN and create dedicated VLAN/SVI to establish a Layer 3 neighborship
● Use vPC peer-link and create dedicated VLAN/SVI to establish a Layer 3 neighborship (least recommended solution)


What exactly is the issue in the least recommended solution? failure domain isolation or lack of?


http://www.cisco.com/c/dam/en/us/td/docs/switches/datacenter/sw/design/vpc_design/vpc_best_practices_design_guide.pdf


Have always gone with separate links in the past but two new factors
- 9K - have the rules changed?
- in the past you had plenty of ports as you're dealing with dozens of 10Gb ports per blade per N7K chassis, now 1/2RUs and suddenly 40Gb/100Gb ports are few as opposed to a blade full of 10Gbs

[burnyd]

Use a separate port. They still follow the same forwarding rules.

[that1guy15]

We have consolidated down from using dedicated links to a single port-channel with SVIs to separate peer-link and non-VPC link.

Either way is fine but why waste the extra ports like you mention. Ive never liked the fact that VPC wants me to carve off multiple 10G interfaces just for the peer-link and its only using 1% or less of the interface bandwidth.

This is on 7K/5Ks but I would assume the 9Ks are pretty similar.