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 14/2/21 19:40, Brian E Carpenter wrote:
> On 15-Feb-21 10:54, Fernando Gont wrote:
>> On 14/2/21 18:34, Ted Lemon wrote:
>>> On Feb 14, 2021, at 4:14 PM, Fernando Gont <[e-mail]
>>> <mailto:[e-mail]>> wrote:
>>>> That's not a legitimate use for GUAs.  Whereas in the case of ULAs,
>>>> that the *intended* usage.
>>>
>>> Untrue. ULAs are intended to be globally unique.
>>
>> Could you compute the birthday paradox for *all ULAs* on the internet,
>> and share the math/results?
> 
> I'm not Ted, and my preferred formulation is:
> 
>     ULAs are intended to be globally unique for all practical purposes.

That's tricky. :-)

ULAs are unique in the topologial span where they are employed.

The thing is that, in *that* topological span, they are unique.

Otherwise, "globally unique for all practical purposes" means "in that 
part of the topology where I care"  (where's "global" implies 
Internet-wide span).



> Although there is a high probability of some ULA prefixes somewhere
> being equal, the probability of your ULA prefix being equal to your
> next-door neighbour's is extremely small.

Doesn't it follow from that that they are intended for some sort of 
org-local, site-local, or admin-defined scope, rather than global?

(same can be said about other prefixes noted by David)


> There's no equivalent statement that can be made for RFC 1918 addresses,
> nor for the deprecated IPv6 Site Local addresses.
> 
> BTW, RFC 4193 seems very clear on this to me:

Agreed on this.

The question is how RFC4193 fits in RFC4007.

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