Re: addition of TLV to locator ID or locator ID set

[Paul Jakma <paul@clubi.ie>]

On Thu, 29 Sep 2005, Brian E Carpenter wrote:

> The reason I think shim6 is intrinsically much more deployable than 
> 8+8 is that it doesn't force *any* change on ULPs. They can go on 
> using the current socket API for a hundred years if they want to.

If shim6 specifically is not to require changes in ULPs:

	|     ULP   |
	|-----------|
	|<interface>|
        |-----------|
        |   shim6   |


Then that means the interface will be either:

- The hosts normal interface between ULP and IP

or

- Network driver interface (so <interface> would actually incorporate
  the hosts IP output layer)

The latter is quite possible given it is intended to have IPSec AH 
and ESP sit above Shim6.

Both of the above interfaces are (obviously) easily encapsulated in 
IP (hey, in the latter case you'll already have a full IP packet 
buffered). In which case there won't be any deep technical barrier to 
just spitting the packet out to another host (eg your normal gateway) 
and letting it to do the shimming.

In which the only thing required to have intermediary shimming (as a 
subset case of normal end-host shim6) really would be to ensure your 
addressing scheme allows for a static 1:1 mapping.

Everything else would be normal shim6, except the hosts would be 
unaware of the shim remapping. Think of the intermediary as being 
exactly like a shim6 host that happens to have a whole /64 allocated 
to it ;).

There ought be no more impact on end-end connectivity than is implied 
by end-host shimming (which AFAIK has "full two-way end-end 
connectivity" as its goal.).

> What has to be deployed is a new IPv6 stack. One host at a time. We 
> know how to do that. It will take years, but it will happen, due to 
> routine operating system upgrades. And there are no discontinuities 
> of service - just progressive growth in the fraction of sessions 
> that can benefit from shim6.
>
> Yes, it is a big change to the stack, but it's *only* the stack.

Intermediary shimming would require an even smaller subset of stacks 
to be upgraded, even more deployable.

regards,
--
Paul Jakma	paul@clubi.ie	paul@jakma.org	Key ID: 64A2FF6A
Fortune:
In any world menu, Canada must be considered the vichyssoise of nations --
it's cold, half-French, and difficult to stir.
		-- Stuart Keate