> Am 24.03.2016 um 22:48 schrieb Sharon Goldberg <[email protected]>:
> 
> Hi Kristof,
> 
> Thanks for your mail.
> 
> I wanted to ask whether you and your team already had a chance to look at the 
> parts of the NTS specification other than the KE protocol.
> 
> No, not yet.
> 
> Specifically:
> - We have specified requirements for the cookie exchange (for which our 
> proposed KE protocol is but one way to solve them). Do you think that if the 
> requirements are met, the rest of NTS' unicast part is solid?
> 
> We can try to look at that in the next short while.
That would be great. Thanks.

> 
> - Also, have you had a chance to look at the broadcast part of the 
> specification?
> 
> Actually we are very interested in broadcast, we even had a few related 
> recent CVEs.  Here is our work describing these CVEs and some suggestions, I 
> hope you will have some time to read it:
> 
> http://www.cs.bu.edu/~goldbe/papers/NTPbroadcast.html 
> <http://www.cs.bu.edu/~goldbe/papers/NTPbroadcast.html>
> 
> One specific suggestion we had was that NTS NOT demobilize ephemeral 
> associations based on crypto errors.  Please read the paper to see why this 
> creates a DoS vulnerability.
> 
> But beyond this, broadcast crypto for NTP is super complex problem.  We have 
> not had time to review this part of the draft, but hope to do so in the 
> future.
> 
> Doing this properly, however, will take some time, and we wanted to focus on 
> the KE for now.  Note that by "properly" I mean with formal proof of security.
> 
> So, one thing we think is needed, and could do, is a formal proof of security 
> for NTS's KE. Using for instance the Canetti Krawcyzk model [1]. We would not 
> feel comfortable claiming that we had reviewed the KE protocol (or the 
> broadcast protocol) until we had done such a proof.  See [2] [3] for examples 
> of what I mean.

Please not that Kristof did an formal analysis of the NTS unicast mode. You 
will find it in [1]. I will send you the paper offline.
[1]     K. Teichel, D. Sibold, and S. Milius, "First Results of a Formal 
Analysis of the Network Time Security Specification," presented at the Security 
Standardisation Research:  Second International Conference, SSR 2015, Tokyo, 
Japan, 2015.

> 
> But, before we started on this proof, which would take some time, we wanted 
> to understand all the requirements. Hence all the questions. Thanks for 
> answering them!
> 
> Sharon
> 
> [1] http://link.springer.com/chapter/10.1007/3-540-44987-6_28#page-1 
> <http://link.springer.com/chapter/10.1007/3-540-44987-6_28#page-1>
> [2] https://eprint.iacr.org/2015/914.pdf 
> <https://eprint.iacr.org/2015/914.pdf>
> [3] http://eprint.iacr.org/2015/978.pdf <http://eprint.iacr.org/2015/978.pdf>
> 
> 
> 
> --
> Sharon Goldberg
> Computer Science, Boston University
> http://www.cs.bu.edu/~goldbe 
> <http://www.cs.bu.edu/~goldbe>_______________________________________________
> ntpwg mailing list
> [email protected]
> http://lists.ntp.org/listinfo/ntpwg

Attachment: signature.asc
Description: Message signed with OpenPGP using GPGMail

_______________________________________________
TICTOC mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/tictoc

Reply via email to