Re: [rtcweb] Questions on draft-ietf-rtcweb-rtp-usage-15

[Colin Perkins <[e-mail]>]

On 17 Jul 2014, at 23:04, Paul Kyzivat <[e-mail]> wrote:
> I have a couple of questions after reading this draft:
> 
> Section 4.9 says:
> 
>   Taking the discussion in Section 11 into account, a WebRTC end-point
>   MUST NOT use more than one RTCP CNAME in the RTP sessions belonging
>   to single RTCPeerConnection (that is, an RTCPeerConnection forms a
>   synchronisation context).
> 
> and then Section 11 says:
> 
>   ... This is motivating the
>   strong recommendation in Section 4.9 to only use a single CNAME.
> 
>      The requirement on using the same CNAME for all SSRCs that
>      originate from the same end-point, does not require a middlebox
>      that forwards traffic from multiple end-points to only use a
>      single CNAME.
> 
> So one is MUST strength, and the other seems to be SHOULD strength. Which is it?
> 
> And isn't this "middlebox" a WebRTC end-point? If so then that is another conflict.

End Point is used in the sense of draft-ietf-avtext-rtp-grouping-taxonomy-02, and I assumed that excluded middleboxes. This is maybe something that should be clarified in the taxonomy.

> Is it really that WebRTC *browsers* must use one CNAME, while other WebRTC *devices* have more freedom? That seems to make sense since an RTCPeerConnection is only a browser thing.

I think that’s essentially what’s meant. Middleboxes can send data with multiple CNAMEs (and have to, in many cases). 

Browsers, and other end systems, ought to use only a single CNAME. Whether that’s SHOULD or MUST use only a single CNAME I don’t have a strong opinion on.

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