Re: [rtcweb] Retransmit: Summary of Alternatives for media keying

[Matthew Kaufman <[e-mail]>]

On Jul 28, 2011, at 1:22 PM, Harald Alvestrand wrote:

> Question that I could probably answer if I understood the DH key exchange thing:
> 
> Is it possible for anyone with packet-replacement access to the media path to perform a MITM attack against DH?

Yes.

> 
> If so: Is it possible to deliver some token by the "high path" (where the SDES keys would go) that ensures that the DH key exchange is with someone possessing that token?

You can do that, or you can check the fingerprints via the high path to authenticate each end and prove that there's no MITM on the media path.

> 
> That would limit MITM attacks to attackers who had access to both the "high path" and the media path.
> 

Correct. And the high path can be protected with HTTPS and the associated PKI.

Matthew Kaufman