Hi Fred,

> -----邮件原件-----
> 发件人: Templin, Fred L [mailto:[email protected]]
> 发送时间: 2011年3月18日 23:17
> 收件人: Xu Xiaohu; [email protected]
> 抄送: [email protected]
> 主题: RE: [Softwires] fwd:
> NewVersionNotificationfordraft-guo-softwire-6rd-ipv6-config-02
> 
> Hi Xiaohu,
> 
> > -----Original Message-----
> > From: [email protected]
> > [mailto:[email protected]] On Behalf Of Xu Xiaohu
> > Sent: Thursday, March 17, 2011 7:04 PM
> > To: Templin, Fred L; [email protected]
> > Cc: [email protected]
> > Subject: Re: [Softwires] fwd: New
> > VersionNotificationfordraft-guo-softwire-6rd-ipv6-config-02
> >
> >
> > > -----邮件原件-----
> > > 发件人: Templin, Fred L [mailto:[email protected]]
> > > 发送时间: 2011年3月18日 1:14
> > > 收件人: Xu Xiaohu; [email protected]
> > > 抄送: [email protected]
> > > 主题: RE: [Softwires] fwd: New Version
> > > Notificationfordraft-guo-softwire-6rd-ipv6-config-02
> > >
> > > Hi Xiaohu,
> > >
> > > I'm just seeing this draft for the first time, and I
> > noticed that it is
> > asking
> > > the CE router to act as a combined DHCP client/relay.
> > Sections 4.1 and
> > > 4.2.3.1 of the VET spec provide a detailed description of
> > how to do that,
> > > and the same considerations apply to what you are trying to
> > accomplish.
> >
> > Hi Fred,
> >
> > Thanks for this information. I will read your draft.
> >
> > > When you start dealing with DHCPv6-allocated addresses and prefixes,
> > > however, you also have to consider how that will interact
> > with the IPv6
> > > routing system. For that, your CEs would either have to engage in a
> > > proactive dynamic IPv6 routing protocol (e.g., RIPng, OSPFv3, etc.)
> > > or somehow discover more-specific IPv6 routes on-demand of
> > data traffic.
> > > Section 5.14 of the VET spec specifies a new on-demand dynamic IPv6
> > > routing protocol based on redirection that applies to
> > ISATAP and VET,
> > > but I think would also apply equally to your use case.
> >
> > This draft (draft-guo-softwire-6rd-ipv6-config-02) only
> > discusses about how
> > to support stateless DHCPv6 service in 6rd scenario. Hence, IMHO, a
> > proactive dynamic IPv6 routing protocol is not needed. Did I
> > misunderstand
> > your point?
> 
> Oh, I may have read too fast; I see now that you are only asking
> about stateless at the current time. But, why not also do stateful
> DHCPv6 prefix delegation (and maybe also stateful DHCPv6 address
> configuration)? A 6rd tunnel should be able to support that the same
> as for an ISATAP tunnel.
> Then, 6rd end sites can use true native IPv6 prefixes instead of the
> IPv4-embedded ones. Should this be added to the draft?

I agree that it should work. However, I wonder what's the advantage of
native IPv6 prefixes over IPv4-embeded ones?

Best wishes,
Xiaohu

> 
> Thanks - Fred
> [email protected]
> 
> > > One question about your draft. If the CE acting as a
> > combined client/relay
> > > forwards a DHCP request to 6rd BR A (via the 6rd anycast
> > address), what
> > > happens if subsequent DHCP renwals are directed to 6rd BR B? I don't
> > > think it is disasterous as long as the BR's communicate
> > with each other,
> > > but you might want to say that somewhere.
> >
> > Since only the stateless DHCPv6 service is now considered in
> > this draft, the
> > synchronization among BRs is not required. Would this usage cause any
> > concern?
> >
> > Best wishes,
> > Xiaohu
> >
> > > Thanks - Fred
> > > [email protected]
> > >
> > > > -----Original Message-----
> > > > From: [email protected]
> > > > [mailto:[email protected]] On Behalf Of Xu Xiaohu
> > > > Sent: Tuesday, March 15, 2011 7:12 PM
> > > > To: [email protected]
> > > > Cc: [email protected]
> > > > Subject: [Softwires] fwd: New Version Notification
> > > > fordraft-guo-softwire-6rd-ipv6-config-02
> > > >
> > > > Hi all,
> > > >
> > > > An updated version of draft-guo-softwire-6rd-ipv6-config is
> > > > available at
> > > > http://tools.ietf.org/html/draft-guo-softwire-6rd-ipv6-config-02.
> > > >
> > > > Abstract:
> > > > The 6rd [RFC5969] linktype does not support IPv6 link-local
> > > > addressing, multicast and 6rd nodes are off-link from each other.
> > > > The host configuration protocol DHCPv6 [RFC3315] relies
> > on link-local
> > > > addressing and multicast to function.  This document specifies how
> > > > DHCPv6 can be used across a 6rd link.
> > > >
> > > > Any comments are welcome.
> > > >
> > > > Best wishes,
> > > > Xiaohu
> > > >
> > > > -----邮件原件-----
> > > > 发件人: IETF I-D Submission Tool [mailto:[email protected]]
> > > > 发送时间: 2011年3月14日 17:45
> > > > 收件人: [email protected]
> > > > 抄送: [email protected]; [email protected]; [email protected]
> > > > 主题: New Version Notification for
> > draft-guo-softwire-6rd-ipv6-config-02
> > > >
> > > >
> > > > A new version of I-D,
> > > > draft-guo-softwire-6rd-ipv6-config-02.txt has been
> > > > successfully submitted by Ole Troan and posted to the
> > IETF repository.
> > > >
> > > > Filename:        draft-guo-softwire-6rd-ipv6-config
> > > > Revision:        02
> > > > Title:           IPv6 Host Configuration in 6rd
> > > > Creation_date:   2011-03-14
> > > > WG ID:           Independent Submission
> > > > Number_of_pages: 5
> > > >
> > > > Abstract:
> > > > The 6rd [RFC5969] linktype does not support IPv6 link-local
> > > > addressing, multicast and 6rd nodes are off-link from each other.
> > > > The host configuration protocol DHCPv6 [RFC3315] relies
> > on link-local
> > > > addressing and multicast to function.  This document specifies how
> > > > DHCPv6 can be used across a 6rd link.
> > > >
> > > >
> > > >
> > > >
> > > > The IETF Secretariat.
> > > >
> > > >
> > > > _______________________________________________
> > > > Softwires mailing list
> > > > [email protected]
> > > > https://www.ietf.org/mailman/listinfo/softwires
> > > >
> >
> > _______________________________________________
> > Softwires mailing list
> > [email protected]
> > https://www.ietf.org/mailman/listinfo/softwires
> >

_______________________________________________
Softwires mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/softwires

Reply via email to