[cabfpub] Verification of Domain Contact and Domain Authorization Document

Wed Jan 3 21:48:01 UTC 2018

Notwithstanding potential discussions to revamp this method, I stand by removal at this time as it is currently dreadfully insecure and nowhere near equivalent to the other methods.

From: Public [mailto:public-bounces at cabforum.org] On Behalf Of Doug Beattie via Public
Sent: Wednesday, January 3, 2018 2:30 PM
To: Ryan Sleevi <sleevi at google.com>; CA/Browser Forum Public Discussion List <public at cabforum.org>; Kirk Hall <Kirk.Hall at entrustdatacard.com>
Subject: Re: [cabfpub] Verification of Domain Contact and Domain Authorization Document

I think this is exactly the type of change that the validation working group should hash out and then propose a solution to the public list.  I’m actually surprised that Tim sent this out given this was on our agenda for tomorrow and he informally polled a number of us on the use of method 1.

I believe that with some additional checks, we can make method 1 sufficiently secure.

From: Public [mailto:public-bounces at cabforum.org] On Behalf Of Ryan Sleevi via Public
Sent: Wednesday, January 3, 2018 3:17 PM
To: Kirk Hall <Kirk.Hall at entrustdatacard.com <mailto:Kirk.Hall at entrustdatacard.com> >; CA/Browser Forum Public Discussion List <public at cabforum.org <mailto:public at cabforum.org> >
Subject: Re: [cabfpub] Verification of Domain Contact and Domain Authorization Document

Given the impact of this, while I don't suggest that the VWG shouldn't take this up, I also don't think that should be reason not to continue the discussion on the public list.

While I understand that Mads and Adriano have suggested they see value in 3.2.2.4.1, I do not think there's a sufficiently compelling demonstration that it remotely approaches the level of assurance of the other methods, or that it fundamentally can. Given that, I do not think 3.2.2.4.1 should be considered - even in modified form - to be equivalent.

In considering how one might theorhetically reform 3.2.2.4.1, I would suggest it's incumbent upon those supporting it to demonstrate how it could achieve that level of assurance. Barring that, we are much better as an industry and a Forum from removing 3.2.2.4.1 unless and until such time as an appropriate demonstration can be made. This avoids introducing significant security risk to the ecosystem due to the Forum's (rather slow) deliberative process.

On Wed, Jan 3, 2018 at 11:08 AM, Kirk Hall via Public <public at cabforum.org <mailto:public at cabforum.org> > wrote:

Tim H, you are chairing the Validation Working Group now – can the VWG take up possible revisions to BR 3.2.2.4.1 as an agenda item?

From: Public [mailto:public-bounces at cabforum.org <mailto:public-bounces at cabforum.org> ] On Behalf Of Adriano Santoni via Public
Sent: Wednesday, January 3, 2018 5:12 AM
To: public at cabforum.org <mailto:public at cabforum.org> 
Subject: [EXTERNAL]Re: [cabfpub] Verification of Domain Contact and Domain Authorization Document

I also concur with Mads, and would support the addition of more requirements to method 3.2.2.4.1.

I like the solution proposed by Mad, but (if I am not mistaken) there is not a specific Whois record field for that information (org number), and I would avoid inserting that information in a field that's not expressly designed for it.

Other solutions may also work, and would be easier to implement, like e.g. mandating a full Registrant address, in the Whois record, which must be one of the official addresses of the Registrant as found in a QIIS/QGIS (excluding, however, all information sources that just publish self-reported organization information, which cannot be regarded as "qualified" information sources and IMO should not be used in the vetting process), and then the "reliable method of communication" should be one that is found in the matching QIIS/QGIS record.

Not sure about method 3.2.2.4.5, at this time, as I have not yet seen a sufficient discussion on it, and I am not sure how it can effectively be used "as is" to obtain a fraudulent certificate.

Adriano

Il 03/01/2018 13:27, Doug Beattie via Public ha scritto:

I agree with Mads and am also supportive of a ballot that removes 3.2.2.4.5 and adds some more detail to 3.2.2.4.1.

Doug

From: Public [mailto:public-bounces at cabforum.org] On Behalf Of Mads Egil Henriksveen via Public
Sent: Wednesday, January 3, 2018 7:11 AM
To: Jeremy Rowley  <mailto:jeremy.rowley at digicert.com> <jeremy.rowley at digicert.com>; CA/Browser Forum Public Discussion List  <mailto:public at cabforum.org> <public at cabforum.org>; geoffk at apple.com <mailto:geoffk at apple.com> 
Subject: Re: [cabfpub] Verification of Domain Contact and Domain Authorization Document

Then I think we should change the requirements. 

As a representative for a CA with a background in strong identity validation (both for natural and legal persons) I find these examples from Ryan and Jeremy to represent a very bad practice. If this really reflects the current practice in the industry, we need to tighten up the requirements and make them much more specific.

>From my point of view (and with my background) I find method 3.2.2.4.1 useful. We must remember that the domain validation methods also are used for EV (and not only OV) and when we have a strongly validated and verified organization (e.g. based on the EV requirements) it makes sense to allow for the organization to apply for certificates including domain names owned by the organization itself. 

I understand that there are doubts about how to ensure that the organization really owns the domain (like in Jeremy’s example), but it should not be too hard to “strengthen” the link between the applicant and the domain owner in terms of rewriting section 3.2.2.4.1. A match in the organization name only should of course not be allowed. 

In Norway every organization is given a unique organization number by a national authority and in the registry for the TLD=.no domains (see www.norid.no <http://www.norid.no> ) we find this organization number as a part of the domain name registrant information. In such cases, we allow for issuance based on 3.2.2.4.1 if the domain name registrant information exactly match organization information (i.e. by country, organization name and organization number).  I think this is a reasonable use case for method 3.2.2.4.1. 

Personally I am more concerned about the possibility we give to any stakeholder in the ecosystem who takes a role in controlling a domain to get an OV (and EV) certificate based on domain control only. This was discussed also in the F2F meeting in Bilbao last year – see https://cabforum.org/2016/05/25/2016-05/#The-Role-of-Identity-in-TLS-Certificates.

Therefore, I am supportive for a ballot which removes 3.2.2.4.5 and keep 3.2.2.4.1 but strengthen this up to allow for use cases like the one described above.

Regards

Mads  

From: Public [mailto:public-bounces at cabforum.org] On Behalf Of Jeremy Rowley via Public
Sent: onsdag 3. januar 2018 05:47
To: geoffk at apple.com <mailto:geoffk at apple.com> ; CA/Browser Forum Public Discussion List  <mailto:public at cabforum.org> <public at cabforum.org>
Subject: Re: [cabfpub] Verification of Domain Contact and Domain Authorization Document

I disagree. The requirements do not specify that.  All that is required is the name of the applicant was verified under 3.2.2.1 and that the register specify the domain contact is the applicant. If Google, Inc. is specified as the domain contact, no address matching is required.

From: geoffk at apple.com <mailto:geoffk at apple.com>  [mailto:geoffk at apple.com] 
Sent: Tuesday, January 2, 2018 4:34 PM
To: Jeremy Rowley <jeremy.rowley at digicert.com <mailto:jeremy.rowley at digicert.com> >; CA/Browser Forum Public Discussion List <public at cabforum.org <mailto:public at cabforum.org> >
Cc: Ryan Sleevi <sleevi at google.com <mailto:sleevi at google.com> >; Adriano Santoni <adriano.santoni at staff.aruba.it <mailto:adriano.santoni at staff.aruba.it> >
Subject: Re: [cabfpub] Verification of Domain Contact and Domain Authorization Document

On Dec 22, 2017, at 12:09 PM, Jeremy Rowley via Public <public at cabforum.org <mailto:public at cabforum.org> > wrote:

The attack vector is easier than that. 

1.	I use very stringent processes to verify that Google, Inc. is a legit company in Utah.
2.	I verify that Jeremy did indeed incorporate Google, Inc. 
3.	I call Jeremy at the phone listed for Google, Inc., the Utah corporation
4.	The domain information shows Google, Inc. as owning  <http://google.com/> google.com
5.	Certificate issues.

Obviously this would be caught in every CA’s high risk checks, but the point remains valid. Regardless of the expertise and thoroughness of the org check, the specs lack any time between the verified org and the actual domain because orgs are not unique on a global basis.

For item 4, you have to verify that “the Applicant is the Domain Contact”.  Obviously it’s insufficient to just compare names—you must verify every element of the WHOIS contact matches the Applicant, that’s typically name, postal address, phone number, and e-mail.

_______________________________________________
Public mailing list
Public at cabforum.org <mailto:Public at cabforum.org> 
https://cabforum.org/mailman/listinfo/public

_______________________________________________
Public mailing list
Public at cabforum.org <mailto:Public at cabforum.org> 
https://cabforum.org/mailman/listinfo/public

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cabforum.org/pipermail/public/attachments/20180103/c493df5b/attachment-0003.html>