Re: [savi] Adopting draft-bagnulo-savi-fcfs/send

[marcelo bagnulo braun <marcelo at it.uc3m.es>]

Frank,

could you please point out who disagreed with the FCFS principle for 
creating SAVI state?
The disucssion so far was whether we use data packets and/or control 
packets for this, but it is always FCFS.

My understanding of the situation is that the current draft only uses 
data packets and Guang has raised the issue (that i understnad actually 
was Fred who brought this up) that some attacks can be launched between 
the DAD and the first data packet

I agreed and i am ok with using ND packets as well to create state (at 
least for IPv6, not clear about IPv4 though).

Moreover, i also think we should use dhcp messages

So, i don't see too much controversy here, but that we are converging to 
something.

Regards, marcelo


Frank Xia escribió:
> Hi Christian
>
> A little bit curious :).
>
> Why are you VERY POSITIVE?
> Have you any proof beyond SAVI mailing list?
>
> Probably, you can just do a simple poll,
> and  IMO  it looks more convincing :-).
>
> BR
> Frank
>
> ----- Original Message ----- From: "Christian Vogt" 
> <christian.vogt at ericsson.com>
> To: "Frank Xia" <xiayangsong at huawei.com>
> Cc: "Fred Baker" <fred at cisco.com>; "Erik Nordmark" 
> <erik.nordmark at sun.com>; "Marcelo Bagnulo Braun" <marcelo at it.uc3m.es>; 
> "Jianping Wu" <jianping at cernet.edu.cn>; "Jun Bi" 
> <junbi at cernet.edu.cn>; "Ana Kukec" <ana.kukec at fer.hr>; "Bill Fenner" 
> <fenner at fenron.com>; "SAVI Mailing List" <savi at ietf.org>
> Sent: Wednesday, January 07, 2009 10:21 PM
> Subject: Re: [savi] Adopting draft-bagnulo-savi-fcfs/send
>
>
> > Frank -
> >
> > As I assured in my earlier email, those issues that have been raised are
> > valid and will certainly have to be addressed.  This includes your
> > comment regarding the support for pre-configured addresses, as well as
> > the question of whether binding establishment should be based on control
> > or data packets, which I summarized in the email you are referring to.
> > But the issues raised do not render the FCFS principle inappropriate for
> > working group adoption.  On the contrary, based on the feedback from the
> > working group and the discussions I had with working group participants,
> > I am very positive that there is in fact strong support for the FCFS
> > principle. Hence the adoption of draft-bagnulo-savi-fcfs/send.
> >
> > - Christian
> >
> >
> >
> > On Jan 7, 2009, Frank Xia wrote:
> >
> > > Hi Christian
> > >
> > > It seems to be a little bit rush :-).
> > >
> > > I don't think people have enough
> > > time to read your previous email.
> > >
> > > For my understanding,  it is hard
> > > to hard to say "STRONG" support
> > > for the FCFS principle.
> > > At least, I can not  find strong support
> > > through this long discussions in the mailing list.
> > >
> > > IMHO, the decision should be based
> > > rough consensus by WG.  IETF is based
> > > on mailing list.
> > > BR
> > > Frank