Re: Policy 2.7.1: MRSP Issue #207: Require audit statements to provide information about which CA Locations were audited

[Ben Wilson via dev-security-policy]

Thanks, Jeff.  These are useful comments, and I will take them into
consideration in revising our proposal.

On Tue, Jan 12, 2021 at 8:38 AM Jeff Ward via dev-security-policy <
dev-security-policy@lists.mozilla.org> wrote:

> On Sunday, January 3, 2021 at 8:38:05 AM UTC-6, Jeff Ward wrote:
> > On Tuesday, December 15, 2020 at 2:41:10 PM UTC-6, Ben Wilson wrote:
> > > All,
> > >
> > > This email is part of the discussion for the next version of the
> Mozilla
> > > Root Store Policy (MSRP), version 2.7.1, to be published during of
> Q1-2021.
> > >
> > > For audit delays, we currently require that audit statements disclose
> the
> > > locations that were and were not audited, but that requirement has not
> been
> > > incorporated yet into the MRSP. See
> > > https://wiki.mozilla.org/CA/Audit_Statements#Minimum_Expectations.
> That
> > > provision reads as follows:
> > >
> > > Disclose each location (at the state/province level) that was included
> in
> > > the scope of the audit or should have been included in the scope of
> the
> > > audit, whether the inspection was physically carried out in person at
> each
> > > location, and which audit criteria were checked (or not checked) at
> each
> > > location.
> > >
> > > - If the CA has more than one location in the same state/province,
> then
> > > use terminology to clarify the number of facilities in that
> state/province
> > > and whether or not all of them were audited. For example: "Facility 1
> in
> > > Province", "Facility 2 in Province, Facility 3 in Province" *or*
> > > "Primary Facility in Province", "Secondary Facility in Province",
> "Tertiary
> > > Facility in Province".
> > > - The public audit statement does not need to identify the type of
> > > Facility.
> > > - "Facility" includes: data center locations, registration authority
> > > locations, where IT and business process controls of CA operations are
> > > performed, facility hosting an active HSM with CA private keys,
> > > facility or
> > > bank deposit box storing a deactivated and encrypted copy of a
> > > private key.
> > >
> > > It is proposed by Issue #207
> > >  that this language
> > > requiring the disclosure of site locations--audited and unaudited--be
> made
> > > clearly part of the MSRP by reference to the language above.
> > >
> > > A similar method of incorporating by reference has been taken in
> section
> > > 2.4 of the MSRP
> > > <
> https://www.mozilla.org/en-US/about/governance/policies/security-group/certs/policy/#24-incidents>
>
> > > with respect to incident reporting and in section 7.1
> > > <
> https://www.mozilla.org/en-US/about/governance/policies/security-group/certs/policy/#71-inclusions>
>
> > > with requirements for the CA inclusion process.
> > >
> > > It is proposed that we add a new subsection 10 to MRSP section 3.1.4
> > > <
> https://www.mozilla.org/en-US/about/governance/policies/security-group/certs/policy/#314-public-audit-information>
>
> > > that would require that audit documentation disclose the facility site
> > > locations that were, or were not, examined.
> > >
> > > One concern that has been raised previously is that the Baseline
> > > Requirements do not define "facility site location". However, we
> believe
> > > that the language above at
> > > https://wiki.mozilla.org/CA/Audit_Statements#Minimum_Expectations
> > > accomplishes that. We're open to suggestions for re-wording parts of
> it to
> > > make it even better.
> > >
> > > Currently, the audit letter template for WebTrust for CAs references
> the
> > > site location audited (at the level of specificity that is proposed
> > > above). Over this past year, due to COVID, some ETSI attestation
> letters
> > > have also explained which sites were and were not checked. This
> approach
> > > seems to work, and the additional information will be beneficial in
> the
> > > future as we evaluate the security and trust of PKI service providers.
> > >
> > > So, for the page cited above, we intend to move "Minimum Expectations"
> out
> > > from under "Audit Delay" so that it stands separately as a requirement
> for
> > > disclosing the facility site location. Then we will also revise MRSP
> > > section 3.1.4 by inserting a new subsection 10 to require "facility
> site
> > > locations that were, or were not, examined" with a hyperlink to the
> Minimum
> > > Expectations language cited above.
> > >
> > > We look forward to your comments and suggestions.
> > >
> > > Sincerely yours,
> > >
> > > Ben
> > Hi Ben. Happy New Year. I have asked the WebTrust Task Force members to
> provide their comments and Don and I will then provide a more detailed
> response. I wanted to be sure to get each of the major firms' feedback
> before responding.
> >
> > Thank you.
> >
> > Jeff
>
> Ben, Don and I offer the following response, which has been vetted through
> the WebTrust Task Force:
>
> Proposed Requirement
> Disclose each location (at the state/province level) that 

Re: Policy 2.7.1: MRSP Issue #218: Clarify CRL requirements for End Entity Certificates

[Rob Stradling via dev-security-policy]

Hi Ben.

> *A CA technically capable of issuing server certificates MUST ensure that
> the CCADB field "Full CRL Issued By This CA" contains either the URL for
> the full and complete CRL or the URL for the JSON file containing all URLs
> for CRLs that when combined are the equivalent of the full and complete CRL*

As a consumer of this data (crt.sh), I'd much prefer to see "Full CRL Issued By 
This CA" and "the URL for the JSON file" as 2 separate fields in the CCADB.  
CAs would then be expected to fill in one field or the other, but not both.  Is 
that possible?

To ensure that these JSON files can be programmatically parsed, I suggest 
specifying the requirement a bit more strictly.  Something like this:
  "...or the URL for a file that contains only a JSON Array, whose elements are 
URLs of DER encoded CRLs that when combined are the equivalent of a full and 
complete CRL"

> I propose that this new CCADB field be populated in all situations where the 
> CA is enabled for server certificate issuance.

Most Root Certificates are "enabled for server certificate issuance".  
Obviously CAs shouldn't issue leaf certs directly from roots, but nonetheless 
the technical capability does exist.  However, currently CAs can't edit Root 
Certificate records in the CCADB, which makes populating these new field(s) "in 
all situations" rather hard.

Since OneCRL covers revocations of intermediate certs, may I suggest that CAs 
should only be required to populate these new field(s) in intermediate 
certificate records (and not in root certificate records)?

Relatedly, "A CA technically capable of...that the CCADB field" seems wrong.  
CCADB "CA Owner" records don't/won't contain the new field(s).  Similar 
language elsewhere in the policy (section 5.3.2) says "All certificates that 
are capable of being used to..." (rather than "All CAs...").

Technically-constrained intermediate certs don't have to be disclosed to CCADB, 
but "in all situations where the CA is enabled for server certificate issuance" 
clearly includes technically-constrained intermediates.  How would a CA 
populate the "Full CRL Issued By This CA" field for a technically-constrained 
intermediate cert that has (legitimately) not been disclosed to CCADB?


From: dev-security-policy  on 
behalf of Ben Wilson via dev-security-policy 

Sent: 08 January 2021 01:00
To: mozilla-dev-security-policy 
Subject: Policy 2.7.1: MRSP Issue #218: Clarify CRL requirements for End Entity 
Certificates

CAUTION: This email originated from outside of the organization. Do not click 
links or open attachments unless you recognize the sender and know the content 
is safe.


This is the last issue that I have marked for discussion in relation to
version 2.7.1 of the Mozilla Root Store Policy
.
It is identified and discussed in GitHub Issue #218

 for the MRSP.

I will soon update everyone on the status of the other 13 discussion items
already presented, as some of them are in need of revision based on
comments received thus far.

While subsection (b) of section 7.1.2.3 of the Baseline Requirements makes
a cRLDistributionPoint (CDP) in end entity certificates optional, Mozilla
still desires that CRL-based revocation information be available because
CRLite uses CRLs to construct its revocation filters.  (Apple also uses
such CRL information in its certificate validation processes and, as I
understand, is making a similar request of CAs with respect to the new
CCADB field, discussed below.)

While all such CRL information is needed, large CRLs are disfavored because
of the time they take to download and process.  Thus, CAs shard, partition,
or "scope" their CRLs into smaller chunks. Section 5 of RFC 5280 explains,
"Each CRL has a particular scope.  The CRL scope is the set of certificates
that could appear on a given CRL. … A complete CRL lists all unexpired
certificates, within its scope, that have been revoked for one of the
revocation reasons covered by the CRL scope.  A *full and complete CRL*
lists all unexpired certificates issued by a CA that have been revoked for
any reason." (Emphasis 

Re: Policy 2.7.1: MRSP Issue #218: Clarify CRL requirements for End Entity Certificates

[Corey Bonnell via dev-security-policy]

Hi Ben,
A few follow-up questions and comments:

1) What are the expectations regarding availability for such CRLs? Do the 
availability requirements in BR 4.10.2 stand for these CRLs even if such CRL 
pointers are not encoded in end-entity certificates?
2) What is the expectation for populating the CRLDP in end-entity S/MIME 
certificates? If no change in policy for S/MIME end-entity certificates is 
desired, then I think the text should be further qualified with "CAs SHOULD 
place the URL for the associated CRL within the crlDistributionPoints extension 
of issued *server* certificates".

Thanks,
Corey

On Thursday, January 7, 2021 at 8:00:46 PM UTC-5, Ben Wilson wrote:
> This is the last issue that I have marked for discussion in relation to 
> version 2.7.1 of the Mozilla Root Store Policy 
> .
>  
> It is identified and discussed in GitHub Issue #218 
>  for the MRSP. 
> 
> I will soon update everyone on the status of the other 13 discussion items 
> already presented, as some of them are in need of revision based on 
> comments received thus far. 
> 
> While subsection (b) of section 7.1.2.3 of the Baseline Requirements makes 
> a cRLDistributionPoint (CDP) in end entity certificates optional, Mozilla 
> still desires that CRL-based revocation information be available because 
> CRLite uses CRLs to construct its revocation filters. (Apple also uses 
> such CRL information in its certificate validation processes and, as I 
> understand, is making a similar request of CAs with respect to the new 
> CCADB field, discussed below.) 
> 
> While all such CRL information is needed, large CRLs are disfavored because 
> of the time they take to download and process. Thus, CAs shard, partition, 
> or "scope" their CRLs into smaller chunks. Section 5 of RFC 5280 explains, 
> "Each CRL has a particular scope. The CRL scope is the set of certificates 
> that could appear on a given CRL. … A complete CRL lists all unexpired 
> certificates, within its scope, that have been revoked for one of the 
> revocation reasons covered by the CRL scope. A *full and complete CRL* 
> lists all unexpired certificates issued by a CA that have been revoked for 
> any reason." (Emphasis added.) 
> 
> There is a new field in the CCADB for CAs to include information needed for 
> browsers or others to construct a "full and complete CRL", i.e. to gather 
> information from CAs that don't include the CRL path to their "full and 
> complete CRL" in end entity certificates they issue. This new CCADB field 
> is called "Full CRL Issued By This CA" and is located under the heading 
> "Pertaining to Certificates Issued by this CA." Rather than condition the 
> requirement that CAs fill in this information in the CCADB only when they 
> don't include a CDP to a full and complete CRL, I propose that this new 
> CCADB field be populated in all situations where the CA is enabled for 
> server certificate issuance. In cases where the CA shards or partitions its 
> CRL, the CA must provide a JSON-based list of CRLs that when combined are 
> the equivalent of the full and complete CRL. 
> 
> Proposed language to add to section 6 of the Mozilla Root Store Policy is 
> as follows: 
> 
> *CAs SHOULD place the URL for the associated CRL within the 
> crlDistributionPoints extension of issued certificates. A CA MAY omit the 
> crlDistributionPoint extension, if permitted by applicable requirements and 
> policies, such as the Baseline Requirements. * 
> 
> *A CA technically capable of issuing server certificates MUST ensure that 
> the CCADB field "Full CRL Issued By This CA" contains either the URL for 
> the full and complete CRL or the URL for the JSON file containing all URLs 
> for CRLs that when combined are the equivalent of the full and complete CRL* 
> . 
> 
> 
> I look forward to your comments and suggestions. 
> 
> Ben
___
dev-security-policy mailing list
dev-security-policy@lists.mozilla.org
https://lists.mozilla.org/listinfo/dev-security-policy