>Keith Moore wrote:
>>> My point is that I believe that a clean separation should be made
>>>between global addresses and scoped addresses. We fully understand
>>>how globals and link-locals work. All the others are still being
>>>hashed out. If we make this break, the address architecture can
>>>move along the standards track. The (great amount of) additional
>>>work that needs to be done on scoped addresses can be carried out
>>>under the scoped address architecture.
>>
>>
>> actually I'd claim that we don't really understand how link-locals
>> work, at least not from the applications viewpoint. but I enthusiastically
>> support the idea of separating the work on globals from the work
>> on scoped addresses.
>
>I believe we do have a good understanding on how link-locals
>work with utility protocols (ICMP, ND, MLD, routing protocols).
>That is why the LL are needed in the base addressing architecture.
>As for LL and user apps, I can't speak authoratively on the subject.
i believe we have some clues on application consideration to scoped
addresses. we may need to document this:
- for numeric representation, scoped address notation, like fe80::1%if0
is available.
- name-to-address mapping is rather complicated.
libc resolvers can sometimes deal with scoped address, assuming
(1) the API is scope-capable (getaddrinfo/getnameinfo), and
(2) the view of scope is the same, i.e. we don't cross node boundary
(3) all data paths are scope-capable.
for instance, /etc/hosts is ok. DNS/NIS is not ok due to (2) and (3).
if server and client are on different machine view of scope is
different so (2) is not satisfied. even if server is collocated with
client DNS uses 128bit notation (AAAA) so scope info is lost/not
handled so (3) is not satisfied.
here are questions i can't really answer:
- what is the exact relationship between link identifier and
interface names/identifiers (API issue)
for instance, KAME assumes one-by-one mapping and uses interface
identifier as link identifier. but the assumption breaks if
we consider configs like multiple interfaces connected to
single ethernet broadcast domain
- what defines a site
there are theoretical definitions but operationally how?
itojun
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
