Larry,

On Thu, Sep 24, 2015 at 6:18 PM, Larry Kreeger (kreeger) <[email protected]>
wrote:
>
>
> If I can restate what I think you two agree on, it is:  If VXLAN evolves
> independently from VXLAN GPE, then a VXLAN GPE endpoint that understands
> only how to be backward compatible with RFC7348 VXLAN will not be able to
> be compatible with the unknown evolution of RFC748.
>
> For this reason, I would discourage anyone from taking RFC7348 VXLAN on a
> separate evolutionary path from VXLAN than what VXLAN GPE is.  I would
> encourage any IETF work on evolving VXLAN to instead evolve VXLAN GPE
> which has an actual version field to prevent unknown forwarding behavior
> from occurring if a VXLAN GPE version 0 endpoint receives an evolved (next
> version) of VXLAN GPE.
>

It is for that reason that I suggested tightening the text.

Anoop
_______________________________________________
nvo3 mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/nvo3

Reply via email to