Re: [rtcweb] No Interim on SDES at this juncture

[Roman Shpount <[e-mail]>]

On Thu, Jun 20, 2013 at 7:09 PM, Richard Barnes <[e-mail]> wrote:

> The path from end to end media path is not the same as the path from end
> to middle (signaling path).  SRTP in general (without assumptions on key
> management) protects against a passive attacker that is on the media path
> but not on the signaling path.
>

My point is that if I can intercept all the traffic from the end point and
the end point uses HTTP to send the keys I can decode everything sent over
SRTP. This means we have one very insecure point. The whole system is as
secure as the least secure portion of it. Thus SRTP with plain HTTP is
exactly as secure as plain RTP (ie not at all).


> If, in addition, the browser does not expose media keys to JS (as is
> required for SDES), then even an active attacker who hijacks the HTTP
> connection to inject scripts cannot access the media keys


DTLS is only slightly more secure with plain HTTP and no identity, since I
would need to hijack the HTTP session and decode/encode media. A bit more
work but marginally so.

Following this logic, should not this group outlaw WebRTC sessions from not
HTTPS sources?
_____________
Roman Shpount