JINMEI Tatuya / 神明達哉 wrote: > > >>>>> On Mon, 18 Mar 2002 12:57:26 -0500, > >>>>> Brian Haberman <[EMAIL PROTECTED]> said: > > > I agree the text should be updated. To address your points 'c' > > and 'd', it should be pointed out that 2462 was written so that there > > was no dependency upon a particular stateful configuration protocol. > > Operators would simply utilize their protocol of choice (DHCPv6 or foo) > > and 2462 would allow that to be advertised to the hosts. > > I agree. How much a spec should be specific is always a tradeoff (if > it is too generic, implementors and operators will be confused against > so many choices. if it is too specific, we'll kill future > extensions), but in this case, I think it is overspecification to > restrict the possibility to DHCPv6.
That may be, but I don't understand how the host would know how to interpet the 'O' bit if it means "do what the operator has decided is correct." If the hosts must be preconfigured to make sense of the 'O' bit, or the 'M' bit for that matter, then I think we've lost quite a bit of utility. In any event, I read 2462 to imply that DHCPv6 is the intended protocol for stateful autoconfiguration of addresses and/or configuration information, by its reference in section 1, but it's certainly implicit, not explicit. The text, and client behavior, need to be clarified one way or the other. -- ===================================================================== Josh Littlefield Cisco Systems, Inc. [EMAIL PROTECTED] 250 Apollo Drive tel: 978-497-8378 fax: same Chelmsford, MA 01824-3627 -------------------------------------------------------------------- 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] --------------------------------------------------------------------
