Re: Path MTU for draft-ietf-v6ops-mech-v2-00.txt

[Erik Nordmark <Erik.Nordmark@sun.com>]

> As currently worded, the specification could result in harmful levels of 
> fragmentation in
> the network when encapsulators implement only the base specification.

My understanding from the previous discussion on the draft the WG did
not want to mandate the dynamic MTU discovery from encapsulator to 
decapsulator. Hence it is optional in the new draft.

But limiting the size of the packets sent into the tunnel to 1280+20
doesn't cause any more fragmentation than the dynamic scheme.
Essentially the mandatory part of the specification means that
the tunnel over IPv4 will provide the link specific fragmentation
(for the link that is IPv4) that the IPv6 base specification requires
for links that can not support 1280 bytes of IPv6 packets.

Having the encapsulator discover the IPv4 MTU doesn't allow us to reduce the
amount of ressembly that the decapsulator needs to perform.
And I don't think the level of fragmentation is harmful - especially given
that most Internet paths support an MTU of 1500 bytes.

  Erik