Re: [Sip] I-D Action:draft-ietf-sip-dtls-srtp-framework-00.txt

[Flemming Andreasen <[e-mail]>]

The SDP Capability Negotiation part is fine albeit you could argue it 
would be cleaner to have the DTLS-SRTP related attributes encoded as 
attribute capabilities rather than as first-level attributes.

-- Flemming

Francois Audet wrote:
> See below. 
>
>   
>> -----Original Message-----
>> From: [e-mail] [mailto:[e-mail]] 
>> On Behalf Of Jason Fischl
>> Sent: Monday, December 03, 2007 16:07
>> To: Audet, Francois (SC100:3055)
>> Cc: [e-mail]; [e-mail]; 
>> [e-mail]; Flemming Andreasen
>> Subject: Re: [Sip] I-D 
>> Action:draft-ietf-sip-dtls-srtp-framework-00.txt
>>
>> On Nov 13, 2007 11:50 AM, Francois Audet <[e-mail]> wrote:
>>     
>>> (copying Flemming)
>>>
>>> Good draft. Very helpful.
>>>
>>> On the SDP side of things.
>>>
>>> I didn't seen anything in the SDP format that specifically 
>>>       
>> indicated 
>>     
>>> in the offer that you are trying to setup DTLS-SRTP.
>>>
>>>       
>> This is signaled in the SDP with the token 
>> "UDP/TLS/RTP/SAVP". See 
>> http://www.ietf.org/internet-drafts/draft-fischl-mmusic-sdp-dt
>> ls-04.txt
>> for more details.
>>     
>
> Ah.... Can you explicitly say that in the document?
> I see there is a reference to this draft, but the document doesn't say 
> that it is used for that purpose.
>
>   
>>> Is the assumption that if there is no indication besides 
>>>       
>> the RTP/SAVP 
>>     
>>> in the tcap, that DTLS-SRTP may be negotiated, and that 
>>>       
>> therefore we 
>>     
>>> do not need an explicit indication that DTLS-SRTP is supported?
>>>
>>>       
>> More specifically, UDP/TLS/RTP/SAVP. to signal we're sending 
>> SRTP over DTLS.
>>
>>     
>>> Do we think it's good enough? In other words, is RTP/SAVP without 
>>> a=crypto or a=key-mgmt good enough to indicate DTLS-SRTP? 
>>>       
>> Or should we have another attribute?
>>     
>> I think this is good enough.
>>     
>
> Agreed.
>
>   
>>> A side comment is that the example shows usage with 
>>>       
>> Best-Effort SRTP. 
>>     
>>> My take is that it could also be used in "DTLS-SRTP-always" mode by 
>>> having the m-line use UDP/TLS/RTP/AVP instead of RTP/AVP 
>>>       
>> and not using 
>>     
>>> a a=tcap/a=pcfg line. It would probably be worthwile to 
>>>       
>> describe it in the draft.
>>     
>> I can add an example fragment to cover this case.
>>     
>
> That would be very useful.
>
> Thanks.
>
>   

_______________________________________________
Sip mailing list  https://www1.ietf.org/mailman/listinfo/sip
This list is for NEW development of the core SIP Protocol
Use [e-mail] for questions on current sip
Use [e-mail] for new developments on the application of sip