Re: Unreachability detection at tunnels servers WAS: RE: Comments onzeroconf draft

[Radhakrishnan Suryanarayanan <rkrishnan.s@samsung.com>]

I think pekka and other members of WG must provide their opinion to your
question.

----- Original Message ----- 
From: "Karen E. Nielsen (AH/TED)" <karen.e.nielsen@ericsson.com>
To: "'Radhakrishnan Suryanarayanan'" <rkrishnan.s@samsung.com>; "'Brian E
Carpenter'" <brc@zurich.ibm.com>
Cc: "'Pekka Savola'" <pekkas@netcore.fi>; "V6OPS WG" <v6ops@ops.ietf.org>;
<juha.wiljakka@nokia.com>; "'Jeroen Massar'" <jeroen@unfix.org>
Sent: Wednesday, September 15, 2004 1:37 PM
Subject: RE: Unreachability detection at tunnels servers WAS: RE: Comments
onzeroconf draft


> Hi Radhakrisnan,
>
> [snip]
> >
> > If IPv6 NUD is also supported then we can either translate
> > the ICMPv4 error
> > messages (faster) or cache out the NUD entry if address
> > resolution fails.
>
> Note that I am talking about RFC 2461 "NUD-like" mechanism, as
> RFC 2461 NUD is based on the fact that address resolution gives proof
> of reachability. This may not be the case on the tunnel links, e.g. in
> Teredo, 6to4 and Isatap, where address resolution in itself may
> not provide proof of reachability.
>
> > The tunnel server must support "packet too big" translation
> > anyway for IPv6
> > PMTU. so as well have a couple of other error messages
> > translated as well
> > which can be helpful.
> >
> > IMHO, ICMPv4 error message translation & IPv6 NUD both should
> > be supported.
> >
>
> OK, that's clear talk. Thanks.
>
> I am still looking for the general sentiment in the WG
> on this, though.
>
> Any opinions ?
>
> BR, Karen
>
> > Regards
> > Radhakrishnan
> >
>
>