(Changing the subject again...)
>>>>> On Fri, 20 Aug 2004 14:33:52 +0100,
>>>>> Tim Chown <[EMAIL PROTECTED]> said:
>> Following the discussions, it isn't entirely clear to me why we
>> could need to open this issue. I think that there is concensus
>> for keeping it as is (as described in Christian's mail).
>>
>> Am I missing something?
> My impression is that the discussion stems from the newly reached consensus
> (since the original semantics were defined in 2462) that M=1 implies that
> RFC3315 functionality is available and O=1 implies that RFC3736 functionality
> is available. With RFC3736 being a subset of RFC3315, it is thus on first
> glance "odd" that you can have M=1, O=0, when M=1 implies RFC3736 support
> is there, as a subset of RFC3315.
> While specific protocols were not mentioned in 2462, this wasn't such an
> issue, but now that they are (in Section 4), and we know the properties
> and features of each, maybe it is?
Assuming we support the idea that M=1 is
Solicit/Advertise/Request/Reply rather than just "RFC3315", I think we
might be clearer on this in documents.
Do you think this will affect rfc2462bis, too?
For example, it currently says in Section 4 that:
A "managed address
configuration (M)" flag indicates whether hosts can use stateful
autoconfiguration [RFC3315] to obtain addresses. An "other stateful
configuration (O)" flag indicates whether hosts can use stateful
autoconfiguration [RFC3736] to obtain additional information
(excluding addresses).
(note that rfc2462bis clarifies in section 1 why it still calls
the protocol for the O flag "stateful". So we don't have to worry
about the "contradiction")
Do you think it should be revised using exchange types? IMO, the
current wording of rfc2462bis does not necessarily contradict our
consensus on the semantics of M=1 (as I assumed above). In
particular, the text for the M flag seems okay, since the exchange
described in [RFC3315] "to obtain addresses" should be
"Solicit/Advertise/Request/Reply". It might still be better to revise
the text for the O flag a bit, such as:
An "other stateful
configuration (O)" flag indicates whether hosts can use a subset of
[RFC3315], as defined in [RFC3736], to obtain additional information
(excluding addresses).
Is this better? Or should we revise more? Or is there even no need
to revise?
JINMEI, Tatuya
Communication Platform Lab.
Corporate R&D Center, Toshiba Corp.
[EMAIL PROTECTED]
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[EMAIL PROTECTED]
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
