Sam Hartman wrote:
Keith, I've considered your points and continue to disagree. I'm
mostly replying in the interest of judging consensus.
I believe that the primary use cases identified in the MIF BOF are use
cases that are not going to go away. I think that saying "avoid
multiple addresses" is likely to be the same kind of head-in-sand
thinking that caused us to get where we are today with a number of
areas where there is a disconnect between what the market wants and
what we're willing to include in our engineering model.
Please note that there is a subtle but important difference between
saying "avoid multiple interfaces" and "don't do that under any
circumstances". Maybe "avoid" is too strong a word and I should have
used "discourage". I agree that the problems presented by multiple
addresses per host are unlikely to go away. They are being driven by
several independent factors.
At the same time it needs to be made clear that adding addresses to a
host does come with a cost, and (e.g.) network admins who wish to solve
problems by adding prefixes to their networks (whether this be for
multihoming or interconnection with private networks or whatever) need
to consider the impact that this will have on applications.
I still don't think that we need to consider this problem from only one
angle. If we do that we will get a solution that only makes sense from
one angle.
Keith
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf