On 26/09/2017 01:03, Andrew wrote:
On Monday, September 25, 2017 at 3:18:44 PM UTC-5, Matthew Hardeman wrote:
Has there been any serious discussion of the potential benefit of CAA reporting 
for certificate issuance attempts?

I'm aware of what the spec says and the SHOULD language, etc...

I'm not a CA and don't represent one.

I do, however, think that it's easier to get buy-in for changes to CA 
infrastructure when there is a strong showing for cost/benefit relationship.

In a post-CT world, issuances which occur will be easily detected quite 
promptly.

I am unsure of the value of a report issuing for a failed issuance attempt.  "Oh, 
yea, that wasn't me.  Someone's looking to attack."  How does that help?  One should 
always assume that one is under attack.  Perhaps it allows you to identify an internal 
party attempting to get a certificate issued and allows you to work with that party to 
correctly get a certificate issued, but wouldn't that legitimate inside party have 
reached out internally anyway when they encountered a problem?

I'm just not sure I understand the point of the reporting in terms of deriving 
real security value.

I think it behooves the community, in selecting items to advance for mandatory 
compliance within the CA space, to choose the requirements imposed carefully 
and with a view to deriving real objective security value.

The BRs are indeed a bit ambiguous on this point, but my interpretation of this 
statement

CAs MUST document potential issuances that were prevented by a CAA record in 
sufficient detail to provide feedback to the CAB Forum on the circumstances, 
and SHOULD dispatch reports of such issuance requests to the contact(s) 
stipulated in the CAA iodef record(s), if present

is that the reports should only be sent in a situation where a certificate 
_would_ have been issued if not for the CAA records. In other words, all other 
security measures intended to prevent misissuance failed, and if not for the 
presence of the CAA record a valid certificate for your domain would have been 
issued to a potentially malicious actor. In that situation, you'd definitely 
want to be informed so you can investigate what allowed the attacker to get 
that far along in the process.

That's my understanding anyway.


Alternatively, you might want to adjust your CAA record if that was
actually a legitimate request.  This is especially handy if the CA only
communicates a generic "rejected" message to the applicant in your
organization (which might even be the person managing the CAA record and
receiving the report).



Enjoy

Jakob
--
Jakob Bohm, CIO, Partner, WiseMo A/S.  https://www.wisemo.com
Transformervej 29, 2860 Søborg, Denmark.  Direct +45 31 13 16 10
This public discussion message is non-binding and may contain errors.
WiseMo - Remote Service Management for PCs, Phones and Embedded
_______________________________________________
dev-security-policy mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-security-policy

Reply via email to