RE: IPv6 Home Use to stimulate deployment over IPv4-NAT

["Bound, Jim" <Jim.Bound@hp.com>]

I caught that in the rfc too.  Yes we need an updated tunnel rfc or
draft.
Would that be valid work here ?  Chairs ?
My only concern about dhcpv4 is that is more work and process.  Which is
fine but we can use vendor class for now?  But I agree dhcpv4 is a good
mechanism.
But we could have something in interim.  Some I know would like to see
this working in 2004 and deployed.
/jim

 


> -----Original Message-----
> From: Erik Nordmark [mailto:Erik.Nordmark@sun.com] 
> Sent: Friday, February 21, 2003 8:42 AM
> To: Bound, Jim
> Cc: Jeroen Massar; JORDI PALET MARTINEZ; v6ops@ops.ietf.org; 
> Alain Durand; alh-ietf@tndh.net
> Subject: RE: IPv6 Home Use to stimulate deployment over IPv4-NAT
> 
> 
> > If we have tunnel brokers we don 't need teredo right?  
> That's my take 
> > now.
> 
> Agreed in principle.
> 
> Two things though:
> Firstly, for the access router to ISP tunnel config
> something simpler (like a DHCPv4 option) might make sense - 
> depends on what type of authentication the ISP wants to do 
> specifically for the IPv6 tunnel. If it is sufficient for the 
> ISP to check that the IPv4 source is one of its customers 
> then the TSP authentication features and flexibility isn't needed.
> 
> Secondly, 
> I've been told that tunnel broker can work trough NAT but the 
> RFC (RFC 3053) says: 3. Known limitations
> 
>    This mechanism may not work if the user is using private IPv4
>    addresses behind a NAT box.
> 
> Thus I think it would be useful to have a specification on 
> how tunnel broker works across a NAT.
> 
>   Erik
> 
>