WGLC for draft-ietf-behave-multicast-09

I am starting the 3rd WGLC on draft-ietf-behave-multicast-09, "IP Multicast
Requirements for a Network Address (and port) Translator NAT".  The WGLC ends
on September 3.  The previous WGLCs were for -06 and -03 (which also went to
IESG).


Draft-ietf-behave-multicast-09 integrates all of the comments received to
date.  Significant changes from -08 include:

* Consideration of Gorry Fairhurst's comments in Chicago about a multicast
source inside of a multi-homed network; unfortunately the best I could
describe to deal with that problem was the new requirement 5-a:

           "a:  If a network is multihomed, the NATs or the network
                configuration MUST ensure that duplicate instances of
                the multicast data traffic does not appear on the public
                network.  This can be accomplished by network design (an
                access control list) or a protocol between the NATs
                (e.g., the NAT with the higher IP address transmits
                multicast data towards the public network and other NATs
                drop it)."

* a NAT with multiple public interfaces SHOULD pair for multicast in the same
way it pairs for unicast (new requirement 4-c).

* The "application considerations" section has been moved to an appendix, as
it seems outside the scope of a document describing the behavior for NATs.


Differences between -09 and -08:
 
http://tools.ietf.org/rfcdiff?url2=http://tools.ietf.org/id/draft-ietf-behave-
multicast-09.txt
  http://tinyurl.com/yqej5l

Differences between -09 and -06:
 
http://tinyurl.com/create.php?url=http://tools.ietf.org/rfcdiff?url1=http://to
ols.ietf.org/id/draft-ietf-behave-multicast-06.txt&url2=http://tools.ietf.org/
id/draft-ietf-behave-multicast-09.txt
  http://tinyurl.com/yulw5t


Please reply with any comments.
-d