On 24/10/16 00:28, Ryan Sleevi wrote:
<snip>
> Either we accept CT compliance is locally defined - which suggests
> optionality - or it's an unnessecary constraint on policy inconsistent
> with existing 6962 implementations.

RFC6962 and 6962-bis specify the technical mechanisms of CT.  Therefore,
it seems entirely appropriate for these documents to define what
constitutes basic CT compliance.

This doesn't prevent TLS Clients from adding extra/stricter requirements
via their own CT Policies though.  For example, 6962-bis draft 19
section 10.2.5 says...
  "To be considered compliant, a certificate MUST be accompanied by at
   least one valid SCT."
...but section 8.1 recognizes that...
  "TLS clients may have policies related to the above risks requiring
   servers to present multiple SCTs.  For example, at the time of
   writing, Chromium [Chromium.Log.Policy] requires multiple SCTs to
   be presented with EV certificates in order for the EV indicator to
   be shown."

Similarly, I think a future Chrome CT Policy could conceivably state
that Chrome will only accept SCTs that correspond to end-entity certs,
even though 6962-bis specifies the option of SCTs for name-constrained
intermediates.

That said, if you're certain that Chrome won't accept SCTs for
name-constrained intermediates as currently defined in 6962-bis, then
I'd be in favour of moving this feature to the Redaction draft.  Since
we've completed WGLC for 6962-bis, I presume this possible course of
action would be a matter for the WG chairs and/or Area Directors to
consider.

-- 
Rob Stradling
Senior Research & Development Scientist
COMODO - Creating Trust Online

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

Reply via email to