On 1/11/16 2:38 AM, Alessandro Vesely wrote:
Hi Jim,
On Sun 10/Jan/2016 23:27:46 +0100 Jim Fenton wrote:
Below is the announcement of a draft I just submitted that may be of interest
to this list. The approach here is complementary to the other proposals I
have
seen along these lines (e.g., smtp-sts).
Your approach looks rather similar to Courier's "SECURITY" extension than
Strict Transport Security. I think you'd be interested in having a look to
the
former. For example, its provision to increase the requirement level allows
practical use of the extension even in the absence of supporting MUAs.
http://www.courier-mta.org/draft-varshavchik-security-smtpext.txt
(That was implemented in 2001, before DANE, and even before SNI.)
Hadn't heard of this; I'll have a look.
Thoughts, reviews, etc. welcomed.
Neither proposal seems to allow clients to specify a set of root CAs (to be
transmitted along with the envelope). That lack is tantamount to assuming
that
the trust relationship is transitive. Is it, or is it me?
Rather than specify root CAs, REQUIRETLS has an option to require that
certificates be verified via DANE (TLSA). My thought is that a sender
that is concerned about PKI-related attacks probably wants to avoid the
CAs entirely.
-Jim
_______________________________________________
ietf-smtp mailing list
ietf-smtp(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf-smtp