Thanks Christian, Am I right in assuming that when MIKEY is being used it has no impact on the SIP/SDP message content? The MIKEY RFC3830 spec. certainly has no examples of any modified SDP lines.
Does the MIKEY protocol work outside of the SIP call setup messages? That is, solely alongside the media (RTP) server. thanks again, Matt -----Original Message----- From: Christian Stredicke [mailto:[EMAIL PROTECTED] Sent: 09 February 2006 21:58 To: Matthew Gardiner Cc: [email protected] Subject: RE: [Sip-implementors] Schemes for SRTP master key exchange We have implemented SDES *and* MIKEY. We found that a show-stopper for mikey is that it takes too long to answer a call (embedded system!). Answer-after=0 becomes impossible. Therefore, we offer only sdes in our firmware. And sdes is very simple, at least the inline modes. Also saving a lot of code, a lot of trouble shooting, and a lot of embedded memory space. Christian > -----Original Message----- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] On Behalf > Of Matthew Gardiner > Sent: Thursday, February 09, 2006 5:02 PM > To: [email protected] > Subject: [Sip-implementors] Schemes for SRTP master key exchange > > Hi all, > > I am currently investigating how to control secure RTP > sessions originating > from one of our new media server products. My main direction > of research is > in discovering what mechanism we should be using in order > that connected SIP > devices exchange master keys. So far, I have encountered two > orthogonal > bodies of information: > > 1. MIKEY (RFC3830) > 2. SDES (draft-ietf-mmusic-sdescriptions-12.txt) > > >From my interpretation of this, two differing schemes are > being presented, > one (MIKEY) proposing that key exchange is implemented using > an additional > protocol - MIKEY and the another (SDES) which proposes that > SIP/SDP be used > to negotiate/exchange crypto parameters. > > Is my above assessment, i.e. the two different schemes, correct? > > If so, do any of the posters to this group know which of > these schemes is > most likely to be implemented in the majority of SIP devices? > All opinions > would be welcomed, since for our product we'd like to avoid > implementing any > necessary protocols. > > thanks, > > Matthew Gardiner > Software Engineer > Aculab > _______________________________________________ > Sip-implementors mailing list > [email protected] > https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors > > > _______________________________________________ Sip-implementors mailing list [email protected] https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors
