Re: Proposed way forward with the transition mechanisms

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

On Mon, Aug 02, 2004 at 04:26:39PM +0300, Pekka Savola wrote:
> > 
> > Did your analysis indicate what type of configuration support is
> > needed for this, and whether direction connection (e.g. between
> > two hosts on the same LAN) is required?
> 
> (I am not sure if I understand what you meant with direct connection,
> above -- clearly if you're directly connected, you don't need to 
> tunnel to your peer..)

Well, if you're "directly connected" in a mechanism that for an IPv6-only
infrastructure tunnels IPv4 over IPv6, I think the implied question is do 
you tunnel direct to the peer, or always via some (other) tunnel end point 
infrastructure/server/device?   (And this may be wider than the same LAN)
 
> This is currently only pointed to from the unmanaged analysis, case D
> (copied below), which practically seems to say "configured v4-in-v6,
> with a tunnel endpoint discovery mechanism".  So, this, at least,
> seems to rather straightforward and simple.  The expectation is that
> you have an association with the tunnel endpoint.  I think this is
> (luckily!) a relatively narrow requirement.

No, I think it is also pointed to by the enterprise scenarios, Scenario 3,
which states:

 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
               re-structuring an existing network, decides IPv6
               is the basis for most network communication.
               Some IPv4 capable nodes/applications will need
               to communicate over that infrastructure.

This really points to some IPv4-in-IPv6 tunneling capability.   You need 
a mechanism to discover the tunneling service and the tunnel endpoint, 
at least.  One example is DSTM.

Tim