Re: [MMUSIC] 1 Week WGLC for draft-ietf-mmusic-rtsp-nat-evaluation-06

[Ari Keränen <[e-mail]>]

Hi,

I had a quick read over the draft and have a few comments:

3.  Requirements on NAT-Traversal

    1.  Must work for all flavors of NATs, including NATs with address
        and port restricted filtering.

The term used by RFC 4787 is "Address and Port-Dependent Filtering"


3.  Requirements on NAT-Traversal

    The list of feature requirements for RTSP NAT solutions are given
    below:

I guess this should say "RTSP NAT traversal solutions".


        *  Address discovery for NAT traversal should take automatically,
           if possible

I assume this means the discovery of the address assigned by the NAT? 
Maybe say that explicitly, e.g., "Discovery of the address(es) assigned 
by NAT should happen automatically if possible".


4.7.4.  Security Considerations

    An ALG will not work whit deployment of end-to-end RTSP signaling

Typo: whit


4.9.1.  [TURN] Introduction

    On the external side this is
    limited to the source address/port pair of the first packet arriving
    on the binding.  After the first packet has arrived the mapping is
    "locked down" to that address.  Packets from any other source on this
    address will be discarded.

This doesn't sound right. This behavior was changed (eventually into 
using permissions) somewhere back in draft-rosenberg-midcom-turn-06. See 
http://tools.ietf.org/html/rfc5766#section-2.3 for up-to-date behavior. 
Check also steps 5 & 7 in the next section and section 4.9.4 for more 
lock down text.


6.  Comparison of NAT traversal techniques

    C2:  How Robust is the solution to changing NAT behaviors

Since the answer is yes/no, maybe this should be "is the solution robust..."


10.  Informative References

    [STUN-IMPL]
               "Open Source STUN Server and Client, http://
               www.vovida.org/applications/downloads/stun/index.html",
               June 2007.

This site seems to be unavailable nowadays. Better reference would be 
"http://sourceforge.net/projects/stun/". And of course, there are plenty 
of other open source STUN implementations, but likewise there are for 
ICE (PJNATH, libnice, ice4j, etc.) if we want to use that as an argument.


Cheers,
Ari (as individual)