On Mon, Jun 14, 2010 at 11:30 PM, RJ Atkinson <[email protected]> wrote:
> Now, all that said, I don't see why this is suddenly
> confusing to anyone.

Hi, Ran,

Thanks for the re-summary. At least, I didn't say it confusing. It was
only that I misunderstood the possible locality scope of identifier,
the scope being a locator(subnet) not a site.

> That noted, mobility can work fine with a local-scope
> ID, because the session Nonce permits differentiation
> of different nodes using the same (e.g. local-scope)
> ID values.  Implementations of ILNP are required to use
> cryptographically-random Nonce values, so an off-path
> attacker won't be able to predict a nonce value externally.

In this scenario, do you also mean a case when the node moves off its
original subnet around in a given site or even across sites?

I think that's even possible even with such a local ID since

   o EUI-64 is rarely duplicate, and even in such a case
   o the Nonce will distinguish connections with the same value of endpoint ID.

If this should be possible, I'd rather use local ID, for then I could
eliminate one infrastructural element managing/governing/distributing
the global IDs.

-- 
DY
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg

Reply via email to