Evaluation: draft-ietf-dnsext-gss-tsig-06.txt

[hardie@qualcomm.com]

Ted Hardie          [ ]     [  ]       [ x]      [   ]

Hopefully quick to resolve.  As I recall, there were versions
of Microsoft's win2k software out there which did not follow this
draft and which were not going to be updated.  Did they use
this algorithm identifier?  That is, are we going to see two different
kinds of behavior associated with that algorithm identifier, depending
on the age of implementation?

I don't really think that is a show stopper, provided a new client
interacting with an old version fails the negotiation or ends up
in some other well knownn state, but if it doesn't, we might have
to consider recommending a new algorithm identifier.   I think
everyone would rather avoid that, though, so hopefully we won't
have to go there.  A warning to implementors that it is the case
might still be useful, though, just as a "don't check your implementation
for conformance against this".