Arturo,

On Mar 2, 2013, at 1:27 AM, Arturo Servin <[email protected]> wrote:
> As far as I can tell, EID space is not Internet routable IPv6 Unicast space.
> Yes, that is ok, but does it have the same constraints of registration and 
> conservation?

Registration: yes.
Conservation: maybe, but likely in the same way IEEE MAC prefixes need to be 
conserved.

> Is anybody able to use the whole /16
> 
> a piece of it, a /20, /32, /48?

It doesn't have to be contiguous.

> It has to be unique?

Obviously.

> Do we need to register it?

Yes.

> If so, do we need a whois/restwhois?

Yes.

> reverse dns?

Maybe -- it isn't clear to me that reverse DNS has much of a future, regardless 
of EIDs.

> who is going to operate those services? 

Yet to be determined.  I suspect once the set of services that need to be 
provided are established, the IETF could either throw it to IANA to figure out 
how the services will be provided or could do an RFP like they did with the RFC 
editor.  As mentioned in my note to Scott, IANA already provides a somewhat 
similar service in the allocation of PENs.

Can you think of any reason why there should be different allocation policies 
for EIDs based on where you live?

> And the coordination requirements appear to me to be quite different from 
> those for IPv6 routable unicast addressing.
> As are the constraints which drive allocation.
> 
> which are?

EIDs can be a flat namespace. This is fundamentally different than how locators 
must be allocated.

Regards,
-drc

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

Reply via email to