Re: [v6ops] Scope of Unique Local IPv6 Unicast Addresses (Fwd: New Version Notification for draft-gont-6man-ipv6-ula-scope-00.txt)

[Fernando Gont <[e-mail]>]

On 7/1/21 02:22, Michael Richardson wrote:
> 
> Fernando Gont <[e-mail]> wrote:
>      > But this is where we go back to the original question:
>      > * RFC4007 says that global scope addresses are globally unique.
> 
>      > * RFC4193 aims to reduce the collision fo a number of ULA prefixes when
>      > grouped together, but certainly does *not* result in globally-unique
>      > prefixes. Still, RFC4193 claims that ULAs globals.
> 
>      > So from the pov of RFC4193, ULAs are globals. From the pov of RFC4007, they
>      > are not.
> 
>      > Which of the two (RFC4007 vs RFC4193) takes precedence?
> 
> It really doesn't matter, because Global has many terms.

It does. RFC 4193 claims that they have global scope.

Global scope does have a meaning.




> ULAs are globally unique (ideally), but are not globally routable.

THey can't be globally routable, because they can't be globally unique.



> Their lack of routability is not an architectural consideration, but a
> bureaucratic RIR-based concern.  They don't get RPSL, RPKI, whois or reverse DNS.
> 
> {Whether you are convinced of the statistics of ULA-R being unique or not,
> does not change the goal that they be unique}

But this is a key thing:
10-1000 nets with no ULAs -> low probability of colission != global scope!



> RFC4007 defines a Global Scope. (Not Global Routing)
> ULAs have Global Scope, and I see nothing in RFC4007 that contradicts that.

  [RFC4007] defines the scope of an address as:

       "[the] topological span within which the address may be used as a
       unique identifier for an interface or set of interfaces"

    And defines the "global scope" to be used for:

       "uniquely identifying interfaces anywhere in the Internet"


You have 40 bits for the "global ID" -- how many networks before 
collisions?  And, at the global level, do you expet to be above or 
bellow that number of networks? If above, you won't have unique 
prefixes, and hence, per the definition of RFC4007, you don't hable 
global scope.



> Unless, you live in IPv4 land, and think that everything that isn't RFC1918
> must be routable. And I'm sure that you don't think that way.

They *can't* be globally routeable, because they are not globally unique.

Similarly, even if you removed bogon filters etc, you cannot use 
10.0.0.0/8 globally because they mean different things for each of us. 
-- they are not global scope.

ULAs are just bigger (and yes, have a recommended algorithm for 
generating prefixes) such that if you grab a few ULA prefixes together, 
it's unlikely that there will be collisions.

Thanks,
-- 
Fernando Gont
SI6 Networks
e-mail: [e-mail]
PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492