[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: set Global ID field to SHA hash of domain name



Hi Zefram,

On Fri, 12 Sep 2003, Zefram wrote:

> George Gross wrote:
> >	At the risk of triggering another firestorm of pro/con debate, is
> >there any reason why the centrally assigned Global ID defined by
> >hinden-ipv6-global-local-addr-02.txt could not be simply the low-order 40
> >bits of a SHA hash of a domain name? i.e. if you own the domain name, you
> >get the IP-v6 global ID for "free"?
>
> That sounds like a decent way of generating a locally assigned ID, but
> it doesn't have the uniqueness guarantee that is the raison d'etre for
> the central registry.  You Have Missed The Point, as one of my favourite
> games can be provoked into saying.

hmmm... I'm not religous about whether the mechanism gets used locally
only or centrally or both.

For the central registry, could we define a reverse DNS lookup to verify
the hash's uniquesness?

Even if the first reverse DNS probe proved not to be unique, one could
append a well-specified computed string, such as "0", "1", "2", ... to the
domain name on each successive attempt, until the hash's uniqueness was
confirmed.

George

>
> -zefram
>


--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@ietf.org
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------