Re: Enterprise Analysis DSTM Issue

[Tim Chown <tjc@ecs.soton.ac.uk>]

On Tue, Aug 09, 2005 at 02:55:55PM -0700, Fred Baker wrote:
> 
> Oh, by the way, as IPv4 address space allocation in fact approaches  
> impossibility, the deployment of IPv6-only infrastructure is not only  
> predictable, it is unavoidable. At this point, it is very avoidable,  
> and I have a hard time arguing that the complications resulting from  
> it are preferable to the complications of running a dual stack network.

I think for the majority of cases/networks that's true.  But I have spoken
to people recently who believe (conventional) dual-stack adds to security
complexity, and they'd like to run one protocol only.  

RFC4057 (section 3.1) defines the IPv6-only infrastructure scenario as
one of the three base secenarios:

   Scenario 3:   IPv6-only network infrastructure with some IPv4-capable
                 nodes/applications needing to communicate over the IPv6
                 infrastructure.  Enterprise deploying a new network or
                 restructuring an existing network, decides IPv6 is the
                 basis for most network communication.  Some IPv4
                 capable nodes/applications will need to communicate
                 over that infrastructure.

so it seems to be appropriate to cover the potential solutions in the
analysis document.   I know Pekka was reluctant to include that scenario
while he was co-chair, but consensus was there and it was included.
 
My view is that something 'DSTM-like' will be needed for Scenario 3, but
(for whatever reason) we have nothing on standards track that solves that
case yet.   The analysis wording should describe the nature of the solution,
and could cite DSTM as one example.    

> I'm all for lrw, v6tc, or whatever it winds up being called. I was  
> last time around as well. AFAIK, v6tc is *also* not chartered to  
> solve the entire complexity of the end to end problem - it is  
> chartered to come up with a rendezvous protocol for a tunneling  
> solution, and presumably not a dozen mutually incompatible ones. 

Agreed, it was supposed to find one solution.

> From what Mark Townsley and Dave Ward told me, something like DSTM might  
> be part of that, but it has to manage an arbitrary underlying tunnel  
> architecture - MPLS, ATM, L2TP, IP/IP6, GRE6, etc, and perhaps a  
> couple of such strung together end to end - and is therefore probably  
> not exactly DSTM. But DSTM would be a useful input. 

I agree.

> I don't think we should be deploying IPv6-only networks in any form until 
> we have figured out the end to end problem.

If there are overall connectivity concerns with the model, these should be 
discussed of course.  Maybe a short text/draft describing the problem would 
be useful input? :)  Brief text and a couple of example concerns?

-- 
Tim/::1