Last Call Review of draft-ietf-fecframe-dvb-al-fec-
review-ietf-fecframe-dvb-al-fec-secdir-lc-kelly-2010-01-09-00

Request Review of draft-ietf-fecframe-dvb-al-fec
Requested rev. no specific revision (document currently at 04)
Type Last Call Review
Team Security Area Directorate (secdir)
Deadline 2010-01-05
Requested 2009-12-03
Other Reviews
Review State Completed
Reviewer Scott Kelly
Review review-ietf-fecframe-dvb-al-fec-secdir-lc-kelly-2010-01-09
Posted at http://www.ietf.org/mail-archive/web/secdir/current/msg01333.html
Draft last updated 2010-01-09
Review completed: 2010-01-09

Review
review-ietf-fecframe-dvb-al-fec-secdir-lc-kelly-2010-01-09

I have reviewed this document as part of the security directorate's ongoing effort to review all IETF documents being processed by the IESG.  These comments were written primarily for the benefit of the security area directors.  Document editors and WG chairs should treat these comments just like any other last call comments.

This document describes implementation of a forward error correction protocol over RTP using already-defined protocol elements. The protocol was originally defined by an ETSI group.

The security considerations section says, "This specification adds no new security considerations to the DVB-IPTV AL-FEC protocol", which I take to mean that the authors see no way in which the proposed approach changes the security properties of the original ETSI specification. Since the protocol doesn't seem to implement any security features, I guess this is probably correct. Still, it might be better to add some additional commentary such as what is found in the security considerations section of draft-ietf-fecframe-interleaved-fec-scheme-07.txt (or, perhaps point to that and the framework doc). 

Lacking much necessary background in this area, I don't feel qualified to fully evaluate this document. With that deficiency noted, the only possible red flag I saw is that the FEC protocol requires that the SSRC fields of the FEC frames be set to 0, while SRTP requires unique SSRC values for security reasons. With my very limited background, I can't be sure if there is an important security interaction here or not, but it seems worth asking about.

--Scott