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

Re: [rfc2462bis] a minor nit in creation of global addresses

>>>>> On Fri, 10 Dec 2004 14:43:13 -0500, 
>>>>> Vlad Yasevich <vladislav.yasevich@xxxxxx> said:

> I am not sure I very much like the statement: 
>       " and the address is not yet in the list"

> This requires an implementation of Stateless Autoconfig
> to keep track of all addresses assigned to an interface
> (even statically assigned ones)  and not just prefixes
> from which an address is formed.  This seems like a new
> requirement on Stateless Autoconfig.

But RFC2462 already required that a host maintain a list of addresses
containing both autoconfigured and manually-configured addresses:

   A host also maintains a list of addresses together with their
   corresponding lifetimes. The address list contains both
   autoconfigured addresses and those configured manually.

So, I don't think this is a new requirement.

Additionally, RFC2462 noted that host configuration variables were
"conceptual" ones and the actual implementation could be different.
In that sense, the host implementation does not have to implement the
additional note in rfc2462bis using a "list" data structure.

(rfc2462bis will remove the note on the "conceptual" nature due to
clarification/simplification regarding host's variables, but I believe
the main point here still applies and is pretty obvious without the
explicit notice).

Does this address your concern?

					JINMEI, Tatuya
					Communication Platform Lab.
					Corporate R&D Center, Toshiba Corp.

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