(coming back to the root of this discussion...) >>>>> On Wed, 11 Aug 2004 18:17:31 +1000, >>>>> Greg Daley <[EMAIL PROTECTED]> said:
> I think that's one of the issues. > It leads to the idea that M|O = 1 can be used to invoke Information-Request. > So in this case, the policy shouldn't be called M policy > and O policy since either the M or O flag can be used to > invoke Information-Request. > Alternatively, > (where ==> is implies) > If we assume that the O=1 ==> Information Request is available, > and we assume that M=1 ==> Rebind/Renew/Request is available, I now understand that the point is: if we simply use [RFC3315] (or stateful DHCPv6) and [RFC3736] (or stateless DHCPv6) then the relationship between the M/O flags and message exchange types (i.e. Solicit/Advertise/Request/Reply or Information-request/Reply) will be unclear. We should make the M/O document so that the relationship will be clear. In this sense, I agree. And then we have two choices: 1. M=1 => full RFC3315 (i.e., both Solicit/Advertise/Request/Reply and Information-request/Reply) is available O=1 => the RFC3376 subset (i.e. Information-request/Reply) is available 2. M=1 => Solicit/Advertise/Request/Reply is available O=1 => Information-request/Reply is available It seems to me that the choice is a controversial issue in this list. Also, some people believe that in choice 2 the combination of M=1,O=0 is "invalid" (meaning they think it is a bad combination). But I personally think whether it's really a bad idea is also a controversial issue. As I showed in a separate message, I can think of a "valid" scenario where the administrator wants to specify the combination. I must confess I've not fully considered either case (so I reserve the right to change my mind in the future:-), but right now I think I slightly prefer choice 2. The reasons are: - we originally thought (in RFC2462) that the M flag (when ON) indicated that the host (should) use the stateful protocol **for address autoconfiguration**. This should mean the M flag (when ON) indicates Solicit/Advertise/Request/Reply. (i.e, the interpretation of choice 2) (Perhaps Greg intended this as "goals of original flags" in San Diego?) - choice 2 is more powerful if we agree that M=1,O=0 in choice 2 can have a valid scenario. In fact, with choice 2 we can describe all possible scenarios of choice 1, but choice 1 cannot represent the equivalent of M=1,O=0 in choice 2. - (I admit this is a subjective opinion) choice 2 makes the semantics of the two flags less dependent, which I think will make the behavior clearer/simpler. 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 --------------------------------------------------------------------