I have been selected as the General Area Review Team (Gen-ART) reviewer
for this draft (for background on Gen-ART, please see
http://www.alvestrand.no/ietf/gen/art/gen-art-FAQ.html).
Please resolve these comments along with any other Last Call comments
you may receive.

Document: draft-ietf-sipping-ipv6-torture-tests-03.txt
Reviewer: Brian Carpenter
Review Date: 2007-10-01
IETF LC End Date: 2007-10-
IESG Telechat date: (if known)
Summary: Ready except for one tricky point
Comments: 

This is clear and easy to understand. Unfortunately, the
underlying ABNF quoted from RFC 3261 is broken. It's copied
from RFC 2373, which was obsoleted by RFC 3513 (in turn
obsoleted by RFC 4291). The correct ABNF for IPv6 literal 
addresses is in RFC 3986 and is quite different.

The specific issue is that the faulty ABNF can generate
this construct:

[2001:db8:::192.0.2.1]

whereas the correct construct is

[2001:db8::192.0.2.1]

Because of this old ABNF error, it's long been unclear whether
the format with ::: should be tolerated, but the format
with :: certainly must be considered as standard, per RFC 4291.

Ideally, RFC 3261 needs fixing. As far as torture tests go,
please consider adding at least one case testing that *both*
the above constructs are tolerated. Otherwise, I could imagine
highly obscure operational mysteries ahead.

     Brian