Re: [saag] Last Call: <draft-dukhovni-opportunistic-security-01.txt> (Opportunistic Security: some protection most of the time) to Informational RFC

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, Aug 07, 2014 at 03:03:26PM +0000, Viktor Dukhovni wrote:
>   - You seem to want to ensure that opportunistic security should
>     avoid defending against active attacks,

[Here "you" == Stephen K.]

That's my take on Stephen's position.  IIRC it derived from wanting no
UI impact from OS.  But DANE lets you securely discover that you can
authenticate a service, authenticate it, and success/failure *is* the
*only* UI needed in that case -- a UI that already exists.

I.e., OS w/ DANE has no UI impact, and you can't fallback on
unauthenticated encryption when the service can be authenticated.  OS w/
DANE has no downgrade attacks.

The only ways to make OS w/ DANE fail are: compromise a DNS registrar in
the chain, compromise the service, compromise the crypto, or DoS.

Heck, OS w/ TOFU/pinning has similar properties once the peer's keys are
learned/pinned (and with all the security considerations of
TOFU/pinning).  DANE isn't the only option, but DNSSEC's secure NXDOMAIN
functionality makes DANE >> TOFU/pinning.

Therefore OS can provide more than unauthenticated encryption in some
cases.

Nico
-- 





[Index of Archives]     [IETF Annoucements]     [IETF]     [IP Storage]     [Yosemite News]     [Linux SCTP]     [Linux Newbies]     [Fedora Users]