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 18/2/21 20:41, David Farmer wrote:
> 
> 
> On Thu, Feb 18, 2021 at 5:10 PM Fernando Gont <[e-mail] 
> <mailto:[e-mail]>> wrote:
> 
>     On 18/2/21 19:52, David Farmer wrote:
[...]
> 
> I agree there are inconsistencies, how fatal those inconsistencies are, 
> I think we disagree.

FWIW, I'm not arguing that they are fatal. For instance, I don't think 
anybody has died out of his -- although I wouldn't want to tempt the 
devils in this respect, either :-) (local saying, so to speak)

IPv6 is mostly all about addressing. I believe we should be able to have 
clear definitions for these things. Or otherwise, just get rid of 
conflicting definitions.


> I believe adding a single word to the definition in 
> RFC4007 will align RFC4007 and RFC4193,  "uniquely identifying 
> interfaces anywhere *reachable* on the Internet".

This is tricky for two reasons:

1) Reachability depends on a lot of things, including filtering/security 
policies

2) this seems like a circular definition
e.g., link-local addresses are only reachable on the local link, because 
they have a link-local scope.  But, base on your definition, they are 
"link-local scope" because they are reachable only within the local link.

3) With your definition, link-locals could be considered GUAs: they are 
indeed unique anywhere reachable on the Internet (which for LLs happens 
to be"the local link").



> Further, I think when we eventually get around to revising RFC4291, 
> every remaining use of the word "scope" needs to be examined, and 
> probably a reference to the updated definition in RFC4007 or its 
> replacement should be added. Note most of the words "scope" in RFC4291 
> are in reference to local and global IIDs, which basically go away with 
> RFC7136. But even after that cleanup, I think there is some confusion 
> around "scope" and its meaning, at least for the uninitiated reader of 
> RFC4291.

It actually seems to be RFC4007 vs everything else. e.g., RFC4007 is at 
odds with RFC4193, too.

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