On 01/04/2012 05:55, RJ Atkinson wrote:
> Earlier, Doug Barton wrote:
>> On 01/03/2012 11:02, Dan Wing wrote:
>>> If IPv6 hosts don't handle
>>> ICMP packet-too-big of less than 1280, those IPv4/IPv6 translators
>>> won't work with sub-1280 MTU IPv4 paths.
>>
>> ... and this is not a feature because? And no, don't quote the
>> robustness principle. The floor for MTU has been hard-coded since day 1,
>> so anyone who breaks that deserves what they get.
>
> In fact, the MTU floor CHANGED from 576 to 1280 well AFTER day 1,
Technically you're correct of course, but for all practical values of
"day 1" I think my point remains valid.
> Separately, I'm old enough to believe in the robustness
> principle -- because Jon Postel's argument for it was
> (and remains) quite correct.
In the abstract, sure. But to what extent does that argument apply to
people who made a best effort at following the spec and didn't get it
done exactly the way that you did; and to what extent does it apply to
people who say "Screw the spec, I'm gonna do it this way, you sort it out."
Doug
--
You can observe a lot just by watching. -- Yogi Berra
Breadth of IT experience, and depth of knowledge in the DNS.
Yours for the right price. :) http://SupersetSolutions.com/
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
