Re-, Please see inline.
Cheers, Med De : Qi Sun [mailto:[email protected]] Envoyé : lundi 15 avril 2013 11:00 À : BOUCADAIR Mohamed OLNC/OLN Cc : [email protected]; Softwires ([email protected]) Objet : Re: [dhcwg] Adoption call on draft-scskf-dhc-dhcpv4-over-dhcpv6 Dear Med, Please see inline. Except "IPv4 addressing"-related configuration, what additional configuration parameters can be provisioned with DHCPv4 and not with DHCPv6? [Qi] There are over 200 DHCPv4 options that may be required to provide IPv4 services. In the period of transition, [Med] The question is not how many dhcpv4 exist, but which ones are really needed when the host is IPv6-enabled. some service providers and services may remain in IPv4 for some time, while end users may demand these IPv4 services to be delivered over IPv6. The required info may be more than IPv4 addresses (port set). As for examples, IMHO, SIP server option and [Med] This is doable using option 21. The name can be resolved to an IPv6 or an IPv4 address. Nothing prevent from doing so. NTP server option are potential customers for this usage. [Med] This is doable using dhcpv6. Why the use dhpv4-over-dhcpv6 is a must? There may be more, because demands are arising with time going on. [Med] I’m still not seeing any particular demand which cannot be honored using dhcpv6. Why in the future, an IPv6-enabled node will require to retrieve configuration using DHCPv4-over-DHCPv6 and not directly using DHCPv6? [Qi] We are talking about IPv6-enabled node in IPv4/IPv6 transition period, when IPv4 and IPv6 services are coexisted. Using DHCPv4-over-DHCPv6 can guarantee all DHCP-based services to be configured correctly. [Med] What are those services? Why those services cannot be configured using DHCPv6? Before designing a solution, we need understand the requirements and the needs first. Of course, in IPv6 only Internet where there is no demand for IPv4 services, we should use DHCPv6. But DHCPv4-over-DHCPv6 can also work in this case. [Med] … except that dhcpv4-over-dhcpv6 introduces more complexity for not obvious reasons. Best Regards, Qi Cheers, Med >-----Message d'origine----- >De : Qi Sun [mailto:[email protected]<mailto:[email protected]>] >Envoyé : lundi 15 avril 2013 07:56 >À : BOUCADAIR Mohamed OLNC/OLN >Cc : Tomek Mrugalski; [email protected]<mailto:[email protected]>; Softwires >([email protected]<mailto:[email protected]>) >Objet : Re: [dhcwg] Adoption call on draft-scskf-dhc-dhcpv4-over-dhcpv6 > > >Dear Med, > >Thanks for your response. Here is my understanding. > >IMHO, defining options in DHCPv6 seems to introduce less complexity in >protocol. However, we do not know for now which options are needed in the >future. If we have to port DHCPv4 options into DHCPv6, how many times >should we do the work? Like one new option per 2 years, or redefining all >DHCPv4 options into DHCPv6 at one time? The proposed solution tries to >avoid polluting the DHCPv6 option space, which is necessary for transition >in DHCP. > >Another point is that, defining option(s) / container would require the >DHCPv6 engine to handle DHCPv4 options, which would introduce more >complexity to the logic and implementation of end points. The proposed >solution leverages the DHCPv4 engine for DHCPv4 messages / options and >DHCPv6 engine for DHCPv6 messages / options. The end points would >compatible to both DHCPv4 process and DHCPv6 process. This is more suitable >for the transition. > >A third consideration is, in the case of IPv6 transition, the DHCP client >(on the tunnel endpoint) is bound to be modified. We think it better to >modify the client completely for the transition purpose. > >Last but not the least, the proposed solution is based on the DHCPv6 >infrastructure, which has been deployed in operators' networks. > >The DHCPv4 over DHCPv6 mechanism is an architectural development for DHCP >protocol to meet the requirements for configuring IPv4 parameters across >the IPv6 network. We think it is important for the DHC WG to work on it. > >Best Regards, >Qi Sun >(Sorry for resending, try to synchronize with softwire ML...) > >On 2013-4-12, at 下午7:47, ><[email protected]<mailto:[email protected]>> ><[email protected]<mailto:[email protected]>> wrote: > >> Dear all, >> >> (I cc sofwtire as this work is also relevant for the working group.) >> >> I do not support this work as it induces more complexity compared to just >defining adequate option in dhcpv6. Defining one container or few DHCPv6 >options is much more simpler that the proposed approach. >> >> Cheers, >> Med >> >>> -----Message d'origine----- >>> De : [email protected]<mailto:[email protected]> >>> [mailto:[email protected]<mailto:[email protected]>] De la part >de >>> Tomek Mrugalski >>> Envoyé : mercredi 3 avril 2013 19:41 >>> À : [email protected]<mailto:[email protected]> >>> Objet : [dhcwg] Adoption call on draft-scskf-dhc-dhcpv4-over-dhcpv6 >>> >>> Hi all, >>> >>> This draft is a result of the discussion we had during last meeting in >>> Orlando. We have very productive discussion about IPv4 hosts >>> configuration in IPv6-only network. After we reached what looked like a >>> consensus in the meeting room, couple folks volunteered to write down >>> the draft that summarizes that perceived consensus. This draft is a >>> product of this effort. >>> >>> As requested by authors, we are announcing adotion call on >>> draft-scskf-dhc-dhcpv4-over-dhcpv6-01. This call is being initiated to >>> confirm whether there is WG consensus to adopt this work as DHC WG >>> draft. Please state whether or not you're in favor of the adoption by >>> replying to this mail. If you are not in favor, please also state your >>> objections in your response. This adoption call will complete on 2013- >04- >>> 17. >>> >>> Regards, >>> Bernie & Tomek >>> >>> On 03.04.2013 19:20, Marcin Siodelski wrote: >>>> In the name of authors of draft-scskf-dhc-dhcpv4-over-dhcpv6 I am >asking >>>> for adoption of this document by DHC WG. >>>> >>>> The latest version (-01) is available here: >>>> >>>> http://www.ietf.org/internet-drafts/draft-scskf-dhc-dhcpv4-over-dhcpv6- >>> 01.txt >>>> http://tools.ietf.org/html/draft-scskf-dhc-dhcpv4-over-dhcpv6-01 >>>> >>>> This document describes a mechanism whereby IPv4 client obtains its >>>> configuration from the DHCPv4 server separated from the client by IPv6- >>>> only network having DHCPv6 service. This mechanism is frequently >>>> referred as "DHCPv4-over-DHCPv6". >>>> >>>> The DHCPv4 client's request is carried in the newly defined DHCPv6 >>>> messages/option to/from the server and can be sent either through a >>>> relay or directly. >>>> >>>> Cheers, >>>> Marcin >>>> >>>> >>>> >>>> _______________________________________________ >>>> dhcwg mailing list >>>> [email protected]<mailto:[email protected]> >>>> https://www.ietf.org/mailman/listinfo/dhcwg >>>> >>> >>> _______________________________________________ >>> dhcwg mailing list >>> [email protected]<mailto:[email protected]> >>> https://www.ietf.org/mailman/listinfo/dhcwg >> _______________________________________________ >> dhcwg mailing list >> [email protected]<mailto:[email protected]> >> https://www.ietf.org/mailman/listinfo/dhcwg _______________________________________________ dhcwg mailing list [email protected]<mailto:[email protected]> https://www.ietf.org/mailman/listinfo/dhcwg
_______________________________________________ Softwires mailing list [email protected] https://www.ietf.org/mailman/listinfo/softwires
