Re: [secdir] Secdir last call review of draft-ietf-rmcat-video-traffic-model-06

[Colin Perkins <[e-mail]>]

> On 24 Jan 2019, at 19:23, Yoav Nir <[e-mail]> wrote:
> 
> Reviewer: Yoav Nir
> Review result: Has Nits
> 
> I have reviewed this document as part of the security directorate's ongoing
> effort to review all IETF documents being processed by the IESG.  Document
> editors and WG chairs should treat these comments just like any other last call
> comments.
> 
> To quote from the abstract, the document "describes two reference video traffic
> models for evaluating RTP congestion control algorithms". Indeed it does not
> describe any protocol or algorithm that is going to get deployed on the
> Internet, but rather a model for evaluating congestion control algorithm before
> they are standardized or deployed. As such, I would not expect it to have much
> to say on security, either good or bad.
> 
> It is conceivable that a congestion control algorithm would be exploitable by
> an attacker. For example, some pattern of traffic might trigger such an
> algorithm to block or slow down traffic for a victim. It may be a good idea to
> evaluate whether such algorithms are conducive to such attacks. But speculation
> such as this are not related to the draft. This draft is about evaluating
> congestion control algorithms for their effect on video quality and frame rates.
> 
> So what is my nit with this?  Why does the Security Considerations section
> contains what it does?
> 
>   It is important to evaluate RTP-based congestion control schemes
>   using realistic traffic patterns, so as to ensure stable operations
>   of the network.  Therefore, it is RECOMMENDED that candidate RTP-
>   based congestion control algorithms be tested using the video traffic
>   models presented in this draft before wide deployment over the
>   Internet.
> 
> This is interesting, but I don't think it has much to do with security. IMO it
> would be enough to say that this document introduces models for evaluation and
> doesn't have any security implications.  The existing text should go somewhere
> else.

To my mind, the security implication is that the algorithm be tested to demonstrate that it doesn’t cause denial-of-service when operating with realistic traffic. This could be, as you note above, that it disrupts the video application by forcing the sending rate to zero; but it’s also important to check that it doesn’t send overly quickly and congest the network, so denying service to other flows. 

-- 
Colin Perkins
https://csperkins.org/