Re: [DNA] Re: Flash renumbering

[Erik Nordmark <erik.nordmark@sun.com>]

Sathya Narayanan wrote:
> Erik -
> 
> I have a question and a comment.
> 
> What is your take on the likelihood of the number of  prefixes on the
> link being large so that they cannot fit in a single CompleteRA?

Two different answers:
I guess the probability is 1.0 that somebody somewhere will configure
such a link.

I guess that this will be very unlikely in common operations, thus the
probability of a link selected at random will have such a large number
of prefixes could be very small. (As we understand it today, multiple
prefixes help with graceful renumbering, and are used in scalable
multihoming, but I don't today see an argument for this resulting in
more than a handful of prefixes.)

> I think I know why you say this. Anyway, #1 is still needed in the
> draft - we should include this in the issues list.

ok

    Erik