Re: [nvo3] Carrying entropy, was: draft-ietf-nvo3-encap-00 should add considerations of traversing NAPT

[[e-mail] (Dale R. Worley)]

Dan Wing <[e-mail]> writes:
> For both points above:   are there cases where the outer header can't
> provide entropy of the inner headers?  With IPv4, Geneve is carried
> over UDP and the UDP source port works fine.  With IPv6, Geneve is
> carried over UDP and the UDP source port and IPv6 flow label can both
> be used.  Is Geneve carried over something that isn't UDP?  Said
> another way, do we want to burn Geneve option bits (and MTU) with this
> entropy information?

I envision Geneve -- as a *generalized* encapsulation technique -- being
used in a broader class of applications.  E.g., Geneve directly over IP,
or over 802.1, or possibly even Geneve directly over the physical medium
(presumably for point-to-point use).  I've always been in favor of
generalizing mechanisms, as long as nobody is *required* to support its
use in full generality and as long as it doesn't introduce overhead into
systems that don't need the generality.

Hence, in this case, having ways of permitting an operational
environment to put entropy in Geneve headers that don't constrain users
of Geneve in more prosaic circumstances.

Dale