Steve, I consider the consensus of this WG to be "the definitive expert". :-)
FWIW though, I concur with Andrew's attack description and impact characterization. Regarding fixing it: I'd rather nuke the redaction option than add further complexity.
On 09/06/16 16:29, Stephen Kent wrote:
Dmitry, I've been waiting for Rob Stradling to comment on the attack described by Andrew. I consider Rob to be the definitive expert on redacted certs (since he invented the concept) and I want to make sure that Ron concurs with the attack description and the impact characterization before adding it to the threat document. If Rob concurs, then I'll draft text to describe the attack and submit it to the WG list for review, rather than re-issuing the whole doc with this added material for review. Andrew's message noted a potential problem for Monitors dealing with redacted certs. The Monitor spec that I co-authored (draft-kent-trans-monitor-auditor-01.txt) already describes a model in which Monitors are expected to have access to the cert data associated with the Subjects they serve. So, in the case of redacted certs, I think a Monitor serving the Subject(s) of such certs will be able to function as expected. Eran suggested that a Monitor need not have such info and might be on the lookout for suspect certs in general, but I have trouble understanding how that will work, in general. I have not reviewed version 16 of 6962-bis, but I found prior descriptions of the Monitor function to be too vague. So, I'll mention the Monitor problem, as well as the CT-aware browser problem (re SCT matching), in the text I prepare. Steve
-- Rob Stradling Senior Research & Development Scientist COMODO - Creating Trust Online _______________________________________________ Trans mailing list [email protected] https://www.ietf.org/mailman/listinfo/trans
