WebRTC is a framework. HTTP is a framework. SIP is a framework. All of these
can be used in very different ways with different deployments. I don't see RTP
being a somehow special relative to many other protocols the IETF develops from
the point of view of not having a minimal interoperable way of securing it.
So lets be blunt here - this document is about justifying that RTP will not
have any MTI security. I will note that rtp-security-options also does not add
any MTI security requirements to RTP. I'm OK with that. All I am raising is
that was that I don't believe that has IETF consensus based on the question at
the last plenary. If you think it does, now would be a great time to outline
that argument.
Lets just keep in perspective here that what we are talking about is the
protocol use for transporting people voice when the PSTN is replaced in the US
and the position of this RFC, and presumable the IETF if it is published, is
that the IETF is not going to mandate a way to secure this.
On Dec 9, 2013, at 3:59 PM, Colin Perkins <csp(_at_)csperkins(_dot_)org>
wrote:
Cullen,
That is absolutely not the intent of the document. From the Introduction, 2nd
paragraph:
The IETF policy on Strong Security Requirements for IETF Standard
Protocols [RFC3365] (the so-called "Danvers Doctrine") states that
"we MUST implement strong security in all protocols to provide for
the all too frequent day when the protocol comes into widespread use
in the global Internet". The security mechanisms defined for use
with RTP allow these requirements to be met. However, since RTP is a
protocol framework that is suitable for a wide variety of use cases,
there is no single security mechanism that is suitable for every
scenario. This memo outlines why this is the case, and discusses how
users of RTP can meet the requirement for strong security.
You will note that last sentence "This memo...discusses how users of RTP can
meet the requirement for strong security". Experience has shown that just
saying "use SRTP" is not sufficient: SRTP is not suitable to secure all uses
of RTP, and the keying mechanisms needed are wildly different across
different use domains. The draft explains in more detail why this is the
case, and makes the case that we need -- as you suggest -- drafts explaining
how to secure each class of applications using RTP (telephony, WebRTC, IPTV,
etc.). It doesn't actually specify security, that it true. Rather is
references to draft-ietf-avtcore-rtp-security-options (also in IETF last
call) which describes the available security options, and refers to the
drafts that do define mandatory-to-implement security for different
application domains (e.g., the WebRTC security architecture).
If some part of the draft is unclear about this, and can possibly be
interpreted to allow insecure use of RTP, please explain what, and we will
fix it.
Colin
On 9 Dec 2013, at 23:24, Cullen Jennings (fluffy)
<fluffy(_at_)cisco(_dot_)com> wrote:
This document says RTP does not need to specify a security mechanisms and
that's OK. Note this document could says you must do A or B depending on the
situation but it does not. It just says it is ok not to specify anything.
My read of the consensus at last plenary was that the IETF had decided it
was going to stop doing that. I don't care about if this draft is published
as it or not as I think this draft will have zero impact to what is
deployed. I do think it is worth us being consistent on what security we
expect to have in RFC going forward.
All I am asking is the IESG be consistent about how they judge consensus on
this and if they decide to publish it, provide some guidance on when they
think it is fine to not have security and when they think it is not fine.
On Dec 6, 2013, at 4:40 AM, Colin Perkins <csp(_at_)csperkins(_dot_)org>
wrote:
Cullen,
On 5 Dec 2013, at 16:56, Cullen Jennings (fluffy)
<fluffy(_at_)cisco(_dot_)com> wrote:
Given the hum in the IETF plenary at the last IETF, I no longer think this
document represents IETF consensus. Given the hum in RTCWeb working group,
I doubt this represents the consensus of the RAI area either.
I think I would be tempted to resolve this by saying for each different
scenario RTP is used in (SIP, RTSP, Mulitcast etc) exactly how it needs to
be secured and for scenarios not listed such as new usages, what the
requirements are. For something like SIP, having just one way to secure
RTP is much better than having many ways.
I'm confused about your objection, since this draft states that we need to
do exactly as you propose.
Colin
--
Colin Perkins
http://csperkins.org/
_______________________________________________
Audio/Video Transport Core Maintenance
avt(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/avt