CA:CertificatePolicyV2.3: Difference between revisions

From MozillaWiki
Jump to navigation Jump to search
Line 71: Line 71:
* [https://groups.google.com/d/msg/mozilla.dev.security.policy/RmEDMqiff3Y/oTVKfbIMAQAJ Audit Report Timing]
* [https://groups.google.com/d/msg/mozilla.dev.security.policy/RmEDMqiff3Y/oTVKfbIMAQAJ Audit Report Timing]
** The current CA policy does not specify when audit reports are due to Mozilla relative to the end date of the audit period.  It only says that CAs must provide the reports to Mozilla within 30 days of receiving the report from their auditor. For the next version of the CA policy, I suggest that this be remedied.
** The current CA policy does not specify when audit reports are due to Mozilla relative to the end date of the audit period.  It only says that CAs must provide the reports to Mozilla within 30 days of receiving the report from their auditor. For the next version of the CA policy, I suggest that this be remedied.
* [https://groups.google.com/d/msg/mozilla.dev.security.policy/j9ntnYr-5o4/CTpcECsKBQAJ Update transitive disclosure policy to reduce duplication of full CA hierarchies]
** Update section 8 of Mozilla's CA Certificate Inclusion policy to say that transitive disclosure is not required when the subject of the CA-certificate is also the subject of a certificate included directly in the Mozilla trust store.


== Proposed Changes That Need To Be Discussed ==
== Proposed Changes That Need To Be Discussed ==

Revision as of 21:22, 8 February 2016

Changes Under Consideration for Version 2.3 of Mozilla's CA Certificate Policy

The DRAFT of Version 2.3 of Mozilla's CA Certificate Policy is here: http://mozilla.github.io/ca-policy/

Changes Made to DRAFT Version 2.3

This is the list of changes that have been made to the DRAFT of Version 2.3 in github, with public view here: http://mozilla.github.io/ca-policy/

  • (C1) Update BR section numbers to correspond with BR version 1.3 that was published in April. https://cabforum.org/wp-content/uploads/RFC3647_Comparison_Table_for_Baseline_Requirements.pdf
    • Also update the process/policy wiki pages to reference the new section numbers
  • Remove Code Signing Trust Bit
    • Decision: Remove references to code signing from Mozilla's CA Certificate Policy, then (after version 2.3 of the policy is published) turn off all code signing trust bits on root certs included in NSS, and remove any root certs that do not have any trust bits enabled.
    • Original proposals:
      • Approved: (D28) Remove Code Signing trust bits. As of Firefox 38, add-ons are signed using Mozilla's own roots. There doesn't appear to be anyone else using the roots in the NSS root store for Code Signing.
      • Not approved: (D12) For Code Signing certs, add requirement to comply with the CA/Browser Forum's Code Signing BRs.
  • Corrections / Minor Updates
    • (C2) Update item #12 of the Inclusion Policy to refer to a more recent version of the CA/Browser Forum Baseline Requirements. And add "or later" to the BR version number.
    • (C3) Remove "ISO 21188:2006 Public key infrastructure for financial services -- Practices and policy framework;" from item #11 of the Inclusion Policy.
      • (D16) In item #11 of the Inclusion policy, remove ISO 21188:2006...
    • (C4) In the first bullet point of item #9 of the Maintenance Policy remove the "after June 30, 2011" and add MD2 and MD4.
      • Current text: "after June 30, 2011, software published by Mozilla will return an error when a certificate with an MD5-based signature is used;"
      • Proposed new text: "software published by Mozilla will return an error when a certificate with an MD2, MD4, or MD5-based signature is used;"
    • (C5) Update the second bullet point of item #9 of the Maintenance Policy to say that certs < 2048 bits now return an error.
      • Current text: "all end-entity certificates with RSA key sizes smaller than 2048 bits must expire by December 31, 2013;"
      • Proposed new text: "software published by Mozilla will return an error when SSL/TLS or Code Signing certificates have RSA key sizes smaller than 2048 bits."
    • (C6) Delete the third bullet point of item #9 of the Maintenance Policy.
      • Current text to be deleted: "after December 31, 2013, Mozilla will disable or remove all root certificates with RSA key sizes smaller than 2048 bits;"
    • (C7) Update section 11 of the Inclusion Policy to refer to and link to more recent versions of the audit criteria.
      • (D24) ETSI seems to be changing their standards, or at least the numbering of them. TS 101 456 and TS 102 042 -> EN 319 411-4 (which is the 4th section of a wider standard, with the whole thing being numbered EN 319 401, note the 0). Updated every year. This change is part of a process numbered STF 458. Planned for draft publication in April 2014, and final in October 2015. This standard will then shadow CAB Forum requirements, and attempt to incorporate changes within 9 months (requires review by 27 member states). Arno and Inigo are the two people to talk to.
        • Recommendation from ETSI: keep the old and the new references of the ETSI standards to allow a smooth transition. When the time comes (in about roughly 2 years) the reference to the TS 102042 and 101456 may be withdrawn.

Proposed Changes That Have Been Discussed

Approved

The following changes have been discussed in the mozilla.dev.security.policy forum, and need to be made to the DRAFT of version 2.3 of the policy in GitHub.

  • Refer to BRs for Name Constraints Requirement
    • (D23) Simplify item #9 of the Inclusion Policy by using Baseline Requirements #9.7, "Technical Constraints in Subordinate CA Certificates via Name Constraints & EKU".
    • (D2) CA/Browser Forum Baseline Requirements version 1.1.6 added a requirement regarding technically constraining subordinate CA certificates, so item #9 of the Inclusion Policy may refer to the BR for details about how to technically constrain a subordinate CA certificate that can sign SSL certs.
    • Approved Text: 9. We encourage CAs to technically constrain all subordinate CA certificates. For a certificate to be considered technically constrained, the certificate MUST include an Extended Key Usage (EKU) extension specifying all extended key usages that the subordinate CA is authorized to issue certificates for. The anyExtendedKeyUsage KeyPurposeId MUST NOT appear within this extension.
      • If the certificate includes the id-kp-serverAuth extended key usage, then the certificate must be Name Constrained as described in section 7.1.5 of version 1.3 or later of the CA/Browser Forum Baseline Requirements for the Issuance and Management of Publicly-Trusted Certificates.
      • If the certificate includes the id-kp-emailProtection extended key usage, then all end-entity certificates MUST only include e-mail addresses or mailboxes that the issuing CA has confirmed (via technical and/or business controls) that the subordinate CA is authorized to use.

Declined

The following changes have been discussed in the mozilla.dev.security.policy forum, and the proposal was either declined or postponed to be re-considered in a future version of the policy.

Proposed Changes Currently in Discussion

The following changes are currently under discussion in the mozilla.dev.security.policy forum.

Proposed Changes That Need To Be Discussed

The following items will be discussed in regards to version 2.3 of Mozilla's CA Certificate Policy.

General Policy Cleanup

  • (D19) Add statement of purpose as described in the policy framework discussion.
  • (D22) Make it clear in the Enforcement Policy that disablement or removal of a root certificate may be scheduled for a future date, in order to allow for customers to transition off of the hierarchy to be distrusted.
  • Require CAs to publish their CPs and CPSes under one of the following Creative Commons licenses: CC-BY, CC-BY-SA or CC-BY-ND. This is so that there is no legal impediment to their proper storage, scrutiny etc. by relying parties.
  • Require at least 20 bits of entropy in the cert serial number.
    • Change Maintenance Policy section 9 from: "all new end-entity certificates must contain at least 20 bits of unpredictable random data (preferably in the serial number)."
    • to "all new end-entity certificates must contain at least 20 bits of unpredictable random data in the serial number."
  • Remove duplication with the BRs.
    • Inclusion Policy Duplication with the BRs:
      • Section 9 re technical constraints -- already handled via D23.
    • Maintenance Policy Duplication with the BRs:
      • Section 2 re reasons for revocation
      • Section 3 re CRL/OCSP repositories
      • Section 8 re algorithms and key sizes
      • Section 9 re preventing algorithm attacks
    • Enforcement Policy Duplication with the BRs:
      • None

Transparency

  • Add requirements about the CA Community in Salesforce: CAs must maintain their data in the CA Community in Salesforce about:
    • All certificates that are capable of being used to issue new certificates, and which directly or transitively chain to their certificate(s) included in Mozilla’s CA Certificate Program that are not technically constrained as described in section 9 of Mozilla's CA Certificate Inclusion Policy.
    • Revoked certificates that were capable of being used to issue new certificates, and which directly or transitively chain to their certificate(s) included in Mozilla’s CA Certificate Program and were not technically constrained as described in section 9 of Mozilla's CA Certificate Inclusion Policy.
  • (D14) Consider adding "except as permitted under CT" to item #4 of the Inclusion Policy, where it says "duplicate issuer names and serial numbers". Then it becomes: "…duplicate issuer names and serial numbers, except as permitted under CT;”
    • Bugzilla bug 1016587
    • Note that version 1.2.2 of the BRs added Appendix B section 5: For purposes of clarification, a Precertificate, as described in RFC 6962 – Certificate Transparency, shall not be considered to be a “certificate” subject to the requirements of RFC 5280 - Internet X.509 Public Key Infrastructure Certificate and Certificate Revocation List (CRL) Profile under these Baseline Requirements.
  • (D13) Consider requiring CAs to publish all issued SSL certs issued by non-technically-constrained certificates. CT is one possible way to meet this requirement.
    • From m.d.s.policy: Although CT would not have prevented issuance, requiring CT for all certificates would have detected the misissuance much sooner. Maybe Mozilla should be the first to require CT for all certificates?

Accountability

  • (D5) Make it clearer that producing syntactically valid certificates is required. In particular, I think that Mozilla should audit a CA's recently-issued certificates and automatically reject a CA's request for inclusion or membership renewal if there are a non-trivial number of certificates that have the problems mentioned in Mozilla's list of Things for CAs to Fix
  • (D9) Make it very clear that a CA with a root certificate included in Mozilla's program is ultimately responsible for every certificate issued that directly or indirectly chains up to the included certificate. If a CA's subcontractors (RAs, subCAs, etc.) have their own practice documentation, it must be inclusive of the CA's practices.
    • The subcontractors may have their own practices in addition to the practices that the CA's CP/CPS impose on them. And the CA's CP/CPS must impose practices that are in line with Mozilla's CA Certificate Policies and CA/Browser Forums Baseline Requirements (depending on the types of certs the function is capable of issuing).
    • The contracting CA must include documentation about the policies/practices that they require the contracted CA to comply with (i.e. in regards to the BRs and Mozilla policy). So the contracting CA's CPS needs to say that the contracted CA must include those things in their CPS, and be audited accordingly.
    • The subcontractor may have their own audit, but it is the CA's responsibility to ensure proper auditing is happening, and to publicly disclose such audits according to section 10 of Mozilla's CA Certificate Inclusion Policy.
    • The CA is responsible for making sure their subcontractors are acting in accordance with Mozilla's CA Certificate Policy and the BRs, including practices and audits. If it is found that a certificate has been mis-issued in the CA's hierarchy, the CA will be held accountable for the mistake, and the root certificate may be removed according to Mozilla's CA Certificate Enforcement Policy.
  • (D21) Address issue of trustworthy CAs
    • knowingly were involved in active attacks against internet users, e.g. by installing software without user consent
    • have a conflict of interest by being involved in the development or sales of software or hardware designed to perform man-in-the-middle attacks on SSL connections

Revocation

  • (D6) When an OCSP response signing certificate expires before the OCSP responses signed by the certificate expire, multiple websites break, particularly sites that use OCSP stapling. Make it a requirement that every OCSP response must have a nextUpdate field that is before or equal to the notAfter date of the certificate that signs it. This should be easy for CAs to comply with.
  • (D7) Add a requirement that every OCSP response must have a nextUpdate field. This is required to ensure that OCSP stapling works reliably with all (at least most) server and client products.
  • (D8) Add a requirement that the nextUpdate field must be no longer than 72 hours after the thisUpdate field, i.e. that OCSP responses expire within 3 days, for every certificate, for both end-entity certificates and CA certificates.
  • (D10) Add requirement for CAs to send Mozilla revoked intermediate certificates by submitting a bug report into the mozilla.org Bugzilla system, filed against the "CA Certificates" component of the "NSS" product. <https://bugzilla.mozilla.org/enter_bug.cgi?product=NSS&component=CA%20Certificates>
  • (D11) Add requirement for CAs to send Mozilla revoked end-entity certificates under certain circumstances by submitting a bug report into the mozilla.org Bugzilla system, filed against the "CA Certificates" component of the "NSS" product. <https://bugzilla.mozilla.org/enter_bug.cgi?product=NSS&component=CA%20Certificates>
    • When to notify Mozilla
      • e.g. CAs must notify Mozilla of end-entity revocations when an end-entity certificate is revoked due to a technical issue that enabled the certificate to be inappropriately used, such as: Wrong Key Usage, Certificate issued for a domain to somebody that doesn't own/control that domain, Mistakenly set isCA=true
      • The cert was issued or used in a way not consistent with the BRs.
    • Time frames for notifying Mozilla, and time frames for the change to happen in OneCRL
  • (D17) Policy changes regarding Improving Revocation.
  • bug 643916 -- Minimize allowed staleness of OCSP/CRL response info in CA root inclusion policy

Government CAs

  • (D18) Audit equivalencies from Government CAs MUST incorporate the CABF BR into their local audit criteria.
  • (D20) Figure out a way to draw a line to say that certain types of actions cannot be taken by the same company/organization that includes the operation of a publicly trusted CA. Things to consider:
    • Government sanctioned/initiated/controlled projects
    • Legally forced participants
    • Ability for one corporation to completely support a spying system like PRISM functionality (physical lines, access nodes, data center/hosting resources, backbone level Internet controls, DNS etc.)
  • (D25) If a government CA does not use a 3rd-party auditor, then the domains that they can issue for should be constrained.
    • Note that when scrolling through the Auditor column in the included spreadsheet it looks like distinguishing based on the auditing would only currently find one included CA. However, this still may be worth considering, as there are other CAs who have applied for inclusion and have their audits done by internal government organizations.
  • Add text to the policy regarding root certificates that are constrained. i.e. if the root is constrained, then the rest of the CA hierarchy is constrained, even if the subCA certs do not have name constraints. Of course the overarching CP/CPS needs to commit to complying with the BRs, and needs to be complaint with Mozilla policy. But would the subCAs all be considered technically constrained even though they do not have name constraints directly in them.

Remaining Cleanup

Proposed Changes we will NOT make to Mozilla's CA Cert Policy

These items have been considered and discussed in mozilla.dev.security.policy, and will not be added to Mozilla's CA Certificate Policy:

  1. In item #8 of the Maintenance Policy add DSA 2048. -- Discussion result:No, we should not add DSA support to Mozilla's CA Certificate Policy, and mozilla::pkix does not need to support DSA certificates.
Retrieved from "https://wiki.allizom.org/index.php?title=CA:CertificatePolicyV2.3&oldid=1116338"