Hi Yiu, Ted and Ian provided an important clarification on the applicability scope of draft-scskf-*. IMHO, this should be recorded in the draft not only in mailing list archives.
Having an applicability scope section in that draft would be useful. Cheers, Med >-----Message d'origine----- >De : Lee, Yiu [mailto:[email protected]] >Envoyé : lundi 15 avril 2013 21:27 >À : BOUCADAIR Mohamed OLNC/OLN; Qi Sun >Cc : [email protected]; [email protected] >Objet : Re: [Softwires] [dhcwg] Adoption call on draft-scskf-dhc-dhcpv4- >over-dhcpv6 > >Med, > >I agree we can talk more motivations in the draft. However, >draft-scskf-dhc-dhcpv4-over-dhcpv6 discusses a generic specification of >DHCPv4 over DHCPv6. I am not sure what we want to mention specific to MAP >or lw4over6 in this draft. > >Thanks, >Yiu > > > >On 4/15/13 11:52 AM, "[email protected]" ><[email protected]> wrote: > >>Re-, >> >>Thanks for the explanations. >> >>I suggest the document to be updated to reflect the clarifications you >>provided and also the ones provided by Ian and Ted for MAP and Lw4over6 >>cases. These are important inputs. >> >>I withdraw my objection to his document. >> >>Thank you all for your patient explanations. >> >>Cheers, >>Med >> >>>-----Message d'origine----- >>>De : Qi Sun [mailto:[email protected]] >>>Envoyé : lundi 15 avril 2013 17:47 >>>À : BOUCADAIR Mohamed OLNC/OLN >>>Cc : [email protected]; [email protected]; [email protected]; >>>[email protected] >>>Objet : Re: [dhcwg] [Softwires] Adoption call on draft-scskf-dhc-dhcpv4- >>>over-dhcpv6 >>> >>> >>>Dear Med, >>> >>>In MAP-E pure stateless mode, IPv4 address (prefix) and port set are >>>provisioned in MAP Rules as designed. But in MAP-E 1:1 mode and lw4over6 >>>which are (kind of) stateful, it has to take into considerations about >>>the >>>lease time etc. issues. In this case, IMHO, DHCPv4 over DHCPv6 is more >>>suitable for IPv4 related configurations. >>> >>>What's more, DHCPv4-over-DHCPv6 is not only designed to deal with the >>>option issues, but also to handle other architectural problems in >>>transition (as Bernie mentioned in previous mail). So I think >>>DHCPv4-over- >>>DHCPv6 is helpful for the evolvement in DHCP architecture. >>> >>> >>>Best Regards, >>>Qi Sun >>> >>> >>>On 2013-4-15, at 下午11:13, <[email protected]> >>><[email protected]> wrote: >>> >>>> Hi Ian, >>>> >>>> Thanks for the clarification. >>>> I understood from your answer: dhcpv6 will be used for both MAP and >>>lw4over6 and both don't require draft-scskf-* for IP address + port >>>provisioning. >>>> >>>> Given currently no additional dhcpv4 only options is required for any >>>>of >>>the solutions we are discussing in softwire, I do still think it is not >>>justified to take on a solution for a problem which may not exist. >>>> >>>> draft-scskf-* proposal can be revived when there is a real need to >>>support dhcpv4-only options. No? >>>> >>>> Cheers, >>>> Med >>>> >>>>> -----Message d'origine----- >>>>> De : [email protected] [mailto:[email protected]] >>>>> Envoyé : lundi 15 avril 2013 16:56 >>>>> À : BOUCADAIR Mohamed OLNC/OLN; [email protected] >>>>> Cc : [email protected]; [email protected]; [email protected] >>>>> Objet : Re: [Softwires] [dhcwg] Adoption call on >>>>>draft-scskf-dhc-dhcpv4- >>>>> over-dhcpv6 >>>>> >>>>> Hi Med, >>>>> >>>>> It would still work for lw-4o6 and the unified CPE. All of the basic >>>>> params for configuring lw4o6/MAP1:1 can be provisioned through the >>>>> OPTION_MAP_BIND that is proposed in the unified CPE draft over DHCPv6. >>>>> Additional DHCPv4 only options would be done via the DHCPv4oDHCPv6 >>>method >>>>> for both lw4o6 and MAP-E. >>>>> >>>>> We still need to agree on which option will be used for provisioning >>>>>the >>>>> address of the lwAFTR/MAP BR, however. There was some discussion on >>>>>this >>>>> on the SW ML last week, but no conclusion reached. >>>>> >>>>> >>>>> Cheers, >>>>> Ian >>>>> >>>>> On 15/04/2013 16:47, "[email protected]" >>>>> <[email protected]> wrote: >>>>> >>>>>> Re-, >>>>>> >>>>>> Thanks for clarifying Ted. I must admit this is not what I understood >>>>>> when I read draft-scskf-*. >>>>>> >>>>>> Does the same conclusion applies also for lw-4over6? (I'm naively >>>>>> assuming, given the approach defined in draft-ietf-softwire-unified- >>>cpe, >>>>>> the same dhcpv6 to configure MAP will also be used lw-4over6) >>>>>> >>>>>> Cheers, >>>>>> Med >>>>>> >>>>>>> -----Message d'origine----- >>>>>>> De : Ted Lemon [mailto:[email protected]] >>>>>>> Envoyé : lundi 15 avril 2013 16:39 >>>>>>> À : BOUCADAIR Mohamed OLNC/OLN >>>>>>> Cc : Qi Sun; [email protected]; Softwires ([email protected]) >>>>>>> Objet : Re: [dhcwg] Adoption call on draft-scskf-dhc-dhcpv4-over- >>>dhcpv6 >>>>>>> >>>>>>> On Apr 15, 2013, at 10:27 AM, <[email protected]> wrote: >>>>>>>> Are you saying MAP is not a concerned with this draft and dhcpv6 >>>>>>>>can >>>be >>>>>>> used for MAP? >>>>>>> >>>>>>> For configuring the MAP-E prefix and port set, yes. That was the >>>>>>> discussion we had in Softwires in Orlando: cover the easy stuff with >>>>>>> DHCPv6 >>>>>>> (this is the existing DHCPv6 MAP option), and then if someone needs >>>>>>> legacy >>>>>>> IPv4 services or stateful address allocation, do it with >>>>>>>DHCPv4-over- >>>>>>> DHCPv6. >>>>>> >>>>>> _______________________________________________ >>>>>> Softwires mailing list >>>>>> [email protected] >>>>>> https://www.ietf.org/mailman/listinfo/softwires >>>> >>>> _______________________________________________ >>>> dhcwg mailing list >>>> [email protected] >>>> https://www.ietf.org/mailman/listinfo/dhcwg >> >>_______________________________________________ >>Softwires mailing list >>[email protected] >>https://www.ietf.org/mailman/listinfo/softwires _______________________________________________ Softwires mailing list [email protected] https://www.ietf.org/mailman/listinfo/softwires
