[cabfpub] Use of wildcard certificates by cloud operators

Ben Wilson ben at digicert.com
Mon May 5 22:12:53 UTC 2014

I think “phishing” is a type of fraud that comes immediately to mind as a type of fraud, but we didn’t want to miss “other types.”  I know that sometimes I use the phrase “fraud or other illegal conduct” so that I cover other bad acts that are not defined as “fraud”.   So, in the case of these cloud domains, one could argue that they have a higher risk that a victim will be phished, and therefore, a CA would have to handle such requests differently, but I’ll let others say what should be done differently.  


From: public-bounces at cabforum.org [mailto:public-bounces at cabforum.org] On Behalf Of Kelvin Yiu
Sent: Monday, May 05, 2014 2:55 PM
To: ben at digicert.com; 'Ryan Sleevi'
Cc: public at cabforum.org
Subject: Re: [cabfpub] Use of wildcard certificates by cloud operators


That’s a good question. Personally, I think “fraudulent” includes cases where the site tries to impersonate another site and steal private information which would including phishing. However, I noticed the BR sometimes calls out phishing explicitly (such as in the definition of High Risk Certificate Request). 


         … which may include names at higher risk for phishing or other fraudulent usage, …


Is there an intention in the BR to categorize “phishing” as distinct from “other fraudulent usages”?




From: Ben Wilson [mailto:ben at digicert.com] 
Sent: Monday, May 5, 2014 10:41 AM
To: 'Ryan Sleevi'; Kelvin Yiu
Cc: public at cabforum.org
Subject: RE: [cabfpub] Use of wildcard certificates by cloud operators


For clarification on #2, does the word “fraudulent” change anything?  In other words, in your minds, what is the scenario that is considered “fraudulent”? 


From: public-bounces at cabforum.org [mailto:public-bounces at cabforum.org] On Behalf Of Ryan Sleevi
Sent: Monday, May 05, 2014 11:23 AM
To: Kelvin Yiu
Cc: public at cabforum.org
Subject: Re: [cabfpub] Use of wildcard certificates by cloud operators




On Mon, May 5, 2014 at 10:10 AM, Kelvin Yiu <kelviny at exchange.microsoft.com> wrote:

This Netcraft post (http://news.netcraft.com/archives/2014/04/28/phishers-find-microsoft-azure-30-day-trial-irresistible.html) has highlighted some questions for us about wildcard certificate requirements in the BR and how they apply to cloud operators. 


Microsoft Azure customers can create sub domain names under specific MS registered domains (e.g. myservice.azurewebsites.net). Users can access these FQDNs using HTTPS, which is is secured with a wildcard certificate managed by Azure. Customers do not have access to the private key of the wildcard certificate. Azure also maintains a process to monitor for fraudulent activities and will take down such sites. AFAIK, other cloud operators such as Google App Engine and Amazon Web Services uses wildcard certificates in a similar way. 


Here are my questions to the forum:


1.       Section 11.1.3 of the BR explicitly disallow wildcard certificates for registry controlled domains (e.g. *.com). The Mozilla maintained http://publicsuffix.org is cited as an example of a public suffix list where Azure, GAE, and AWS domains can be found. Does the current usage of wildcard certificates by cloud operators violate the BR? If so, is this intentional and what is the reason?

The Public Suffix List is presently divided into two sections - ICANN-assigned suffices (eg: gTLDs, ccTLDs) and "Private" (enterprise) suffices, for which Azure falls under (as do many other hosted services, such as AWS and Google App Engine)


In both cases, things are fuzzy - in a know-it-when-I-see-it sort of way. That is, it's easy to say something like "*.com" should be invalid. For something like *.appspot.com or *.azurewebsites.net, one would say that it should be invalid for "just anyone" to register, but it should be valid for Google/Microsoft (respectively) to register - as they have. So perhaps we'd say this is a "high-risk domain" and requires manual vetting, but it's hard to say as a general rule it should be prohibited.


Further, it gets more troubling with something like *.nike (as a hypothetical). If Nike doesn't allow third-party registrations, but instead fully operates the domain - much like *.appspot.com or *.azurewebsites.net - should it be allowed to register a wildcard cert? Maybe, maybe not. (For the record, Chrome, in present form, would reject it, as it would any wildcard to an ICANN/IANA-assigned gTLD)


Note that Section 11.1.3 provides for registrations in *.azurewebsites.net and *.appspot.com - as well as *.nike - in its present form, in the first sentence of the second paragraph:


"If a wildcard would fall within the label immediately to the left of a registry-controlled† or public suffix, CAs MUST refuse issuance unless the applicant proves its rightful control of the entire Domain Namespace."


2.       Section 13.1.5 of the BR explicitly require wildcard certificates that were “used to authenticate fraudulently misleading subordinate FQDN” to be revoked within 24 hours. If the fraudulent sites never had access to the private key of the wildcard certificate and the cloud operator has a process to take down fraudulent sites, should these wildcard certificates be required to be revoked?


I don't have a good answer for this, and will probably need to think more before responding. I can see positive and negative outcomes with either answer.




Public mailing list
Public at cabforum.org


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cabforum.org/pipermail/public/attachments/20140505/a7d064bb/attachment-0003.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 5453 bytes
Desc: not available
URL: <http://lists.cabforum.org/pipermail/public/attachments/20140505/a7d064bb/attachment-0001.p7s>

More information about the Public mailing list