Hi Ben, RFC 5280, section 5.2.5 <https://datatracker.ietf.org/doc/html/rfc5280#section-5.2.5> (which defines the syntax of the IDP CRL extension) says:
“The syntax and semantics for the distributionPoint field are the same as for the distributionPoint field in the cRLDistributionPoints extension (Section 4.2.1.13 <https://datatracker.ietf.org/doc/html/rfc5280#section-4.2.1.13> ).” Section 4.2.1.13 says: “When the distributionPoint field is present, it contains either a SEQUENCE of general names or a single value, nameRelativeToCRLIssuer. If the DistributionPointName contains multiple values, each name describes a different mechanism to obtain the same CRL.” The issue that John describes requires a violation of the requirements above, as multiple IDP distributionPoint names/URLs within a given CRL must identify that same CRL. If the included names/URLs actually refer to CRLs that differ in scope/revoked entry content, then the “same CRL” rule is not being followed by the CA. Perhaps it would be good to call this out in Mozilla policy, but I believe the CRL profile requirements in RFC 5280 address this concern. Thanks, Corey From: 'Ben Wilson' via CCADB Public <[email protected]> Sent: Wednesday, July 2, 2025 4:48 PM To: Dimitris Zacharopoulos (HARICA) <[email protected]> Cc: Chris Clements <[email protected]>; CCADB Public <[email protected]>; [email protected] <[email protected]>; [email protected] <[email protected]> Subject: FORK: Re: Further Improving the CCADB Policy (FEEDBACK REQUESTED) Hi All, Section 6.2 of CCADB Policy (for the JSON Array of Partitioned CRL URLs) says, "CA Owners MUST ensure that each corresponding CRL contains a critical ‘Issuing Distribution Point’ extension and the ‘distributionPoint’ field of the extension MUST include a ‘UniformResourceIdentifier’. The value of the UniformResourceIdentifier MUST exactly match a URL, from which the CRL was accessed, present in the CCADB record associated with the CA certificate." John Schanck has pointed out that this language might be too permissive because "`distributionPoint` field can contain multiple URIs, so "match a URL [...] present in the CCADB record" implies that a CA could comply with this clause by listing each of the entries of their "JSON Array of Partitioned CRL URLs" in the `distributionPoint` field. A CRL with the full list of partitioned CRL URLs in its idp extension could be substituted for any other CRL in the list." He is suggesting that the language be revised to read, "CA Owners MUST ensure that each corresponding CRL contains a critical 'Issuing Distribution Point' extension. The 'distributionPoint' field of that extension MUST include exactly one 'UniformResourceIdentifier' that matches a URL present in the CCADB record associated with the CA certificate. The extension MAY include other UniformResourceIdentifiers." Are there any suggestions, recommendations, or improvements to address these concerns? Ben On Tue, Jul 1, 2025 at 10:19 AM Dimitris Zacharopoulos (HARICA) <[email protected] <mailto:[email protected]> > wrote: On 27/6/2025 1:18 π.μ., Chris Clements wrote: Hi Dimitris, It seems switching between full and partitioned CRLs is acceptable so long as at least one of the two corresponding fields in the CCADB is accurately populated at all times. If there are multiple full CRL URLs during a period of transition, at least one of those will need to be disclosed to the CCADB. With that said, is there a suggested wording change for the CCADB Policy? Hi Chris, I believe my biggest concern comes from the recent change introducing the following sentence: * "MUST match exactly as they appear in the certificates issued by the corresponding CA" Since a CA may change URLs arbitrarily, it is clear that this rule will fail because there will be a moment in time where an early-issued end-entity certificate will have "URL A", and a later-issued certificate will have "URL B". In that case, CCADB will have either "URL A" or "URL B" causing issues with the policy. I suggest removing this sentence. If I recall correctly, the purpose of this field was to offer an out-of-band CRL checking mechanism that Certificate Consumers may use in addition to the CRLDP URL in the actual certificates, or the lack of such an extension. To assist with the interchange of full and partitioned URLs, I suggest removing the following two sentences because as described, the actual implementation is "either" and not "either or both": * If populating a full and complete CRL URL: * the JSON Array of Partitioned CRL URLs field MUST be empty. * If populating a JSON Array of Partitioned CRL URLs: * the full and complete CRL URL MUST be empty. Best regards, Dimitris. Thank you -Chris On Thursday, June 26, 2025 at 1:37:02 PM UTC-4 [email protected] <mailto:[email protected]> wrote: This is the same for Apple. Whatever is populated will be picked up, so as long as at least one of the fields is populated at all times CAs can switch between full and partitioned CRLs as they’d like (from Apple’s perspective). Cheers, -Clint On Jun 26, 2025, at 10:27 AM, 'Ben Wilson' via CCADB Public <[email protected] <mailto:[email protected]> > wrote: For CRLite, populating both fields should not create a problem. However, CAs should not populate the "Full CRL" field with more than one URL. Thanks, Ben On Thu, Jun 26, 2025 at 10:56 AM 'Aaron Poulsen' via CCADB Public <[email protected] <mailto:[email protected]> > wrote: I am also interested to know if a process exists for transitioning from a full to partitioned CRL (or the reverse) in CCADB. I can't find any discussions that address this need. CCADB states that at least one field must be populated. Can we simply provide values for both fields? Aaron On Monday, June 23, 2025 at 2:42:46 AM UTC-6 Dimitris Zacharopoulos (HARICA) wrote: Hi Chris, Regarding the URLs of CRLDPs, it is non uncommon for a CA to change the URL of the CRLDP. Obviously, there are redirects until the certificates with the old URL expire. What is the expectation if there are multiple URLs to be included in CCADB? Also, what is the process for switching from sharded to full CRLs and vice-versa? Best regards, Dimitris. On 17/6/2025 3:45 μ.μ., 'Chris Clements' via CCADB Public wrote: All, The updated CCADB <https://www.ccadb.org/policy> Policy and <https://www.ccadb.org/cas/incident-report> Incident Reporting Guidelines have been published with an effective date of July 15, 2025. This date is in line with the expectation that CA Owners will have updated their in-use TLS server authentication certificate validation methods for publicly-trusted hierarchies, following the <https://groups.google.com/a/ccadb.org/g/public/c/QOFzi8wGf8Y/m/Tg3ULE0KAgAJ> recent CCADB enhancement. CA Owners are strongly encouraged to align their CCADB disclosures with the expectations that become effective on July 15, 2025, as soon as reasonably possible. Please note, to comply with the updated CRL disclosure requirements described in <https://www.ccadb.org/policy#62-certificate-revocation-list-disclosures> Section 6.2, it is expected some CAs will need to update existing certificate records (e.g., modifying CRL disclosures to exactly match those included in certificates). Thank you -Chris, on behalf of the CCADB Steering Committee On Fri, May 30, 2025 at 4:26 PM Chris Clements <[email protected] <mailto:[email protected]> > wrote: All, Thank you to everyone who provided valuable feedback on the proposed <https://github.com/mozilla/www.ccadb.org/pull/198> updates to the CCADB Policy and the Incident Reporting Guidelines (IRGs). Both artifacts have been enhanced thanks to the insightful recommendations and suggestions. We want to reiterate the original objectives for this update: * Clarifying Root Store Operator expectations for CCADB disclosures. * Streamlining requirements across different root programs to reduce redundancy. * Enhancing the simplicity and readability of the policy. We plan to publish the updated CCADB Policy and IRGs later in June, with an effective date of July 1, 2025. We appreciate your continued collaboration in making the CCADB a more effective and transparent resource for the community. -Chris, on behalf of the CCADB Steering Committee On Fri, May 2, 2025 at 10:48 AM Chris Clements <[email protected] <mailto:[email protected]> > wrote: All, Following the community’s recent iteration <https://groups.google.com/a/ccadb.org/g/public/c/GIBHz9FUjHY/m/XOOLNpOFCAAJ> and improvement on the updated CCADB Incident Reporting Guidelines <https://www.ccadb.org/cas/incident-report> (IRGs), the CCADB Steering Committee has collaborated on an updated draft of the CCADB Policy. The set of proposed updates are available here <https://github.com/mozilla/www.ccadb.org/pull/198> . Objectives for this update include: * Clarifying Root Store Operator expectations related to CCADB disclosures. Some of these clarifications were motivated by public Incident Reports filed to Bugzilla over the past year. * Creating opportunities to (a) remove redundant/similar requirements located across root program policies related to CCADB disclosures and (b) encourage future simplicity. * Promoting simplicity and improving readability through a reorganization of normative and non-normative requirements. Additionally, minor updates are proposed to the IRGs (e.g., further streamlining the incident reporting closure process). These proposals should not be considered “final”, but instead a “work in-progress” that we hope to enhance through community contributions. We welcome your feedback on these proposed updates and recommendations by May 23, 2025. Please share your thoughts by replying to this email or, preferably, by suggesting edits directly on GitHub. Thank you -Chris, on behalf of the CCADB Steering Committee -- You received this message because you are subscribed to the Google Groups "CCADB Public" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected] <mailto:[email protected]> . To view this discussion visit https://groups.google.com/a/ccadb.org/d/msgid/public/CAAbw9mApWKw1J2ZJ2-1ft7LgJK7CDiYvHKLFjtNEiH1EH%3DmYtQ%40mail.gmail.com <https://groups.google.com/a/ccadb.org/d/msgid/public/CAAbw9mApWKw1J2ZJ2-1ft7LgJK7CDiYvHKLFjtNEiH1EH%3DmYtQ%40mail.gmail.com?utm_medium=email&utm_source=footer> . -- You received this message because you are subscribed to the Google Groups "CCADB Public" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected] <mailto:[email protected]> . To view this discussion visit https://groups.google.com/a/ccadb.org/d/msgid/public/48947bbc-ca5d-4922-a9fd-9bb74fd82bf9n%40ccadb.org <https://groups.google.com/a/ccadb.org/d/msgid/public/48947bbc-ca5d-4922-a9fd-9bb74fd82bf9n%40ccadb.org?utm_medium=email&utm_source=footer> . -- You received this message because you are subscribed to the Google Groups "CCADB Public" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected] <mailto:[email protected]> . To view this discussion visit https://groups.google.com/a/ccadb.org/d/msgid/public/CA%2B1gtaZh0hjfGtc_6aKeF2vJtQqRh3SLpkTeSdiipDzePVgkOg%40mail.gmail.com <https://groups.google.com/a/ccadb.org/d/msgid/public/CA%2B1gtaZh0hjfGtc_6aKeF2vJtQqRh3SLpkTeSdiipDzePVgkOg%40mail.gmail.com?utm_medium=email&utm_source=footer> . -- You received this message because you are subscribed to the Google Groups "CCADB Public" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected] <mailto:[email protected]> . To view this discussion visit https://groups.google.com/a/ccadb.org/d/msgid/public/CA%2B1gtaZ8TXSqnG_O697Qmc0ki7L-o20jBrZbiKq_LaMeo-e3Uw%40mail.gmail.com <https://groups.google.com/a/ccadb.org/d/msgid/public/CA%2B1gtaZ8TXSqnG_O697Qmc0ki7L-o20jBrZbiKq_LaMeo-e3Uw%40mail.gmail.com?utm_medium=email&utm_source=footer> . -- You received this message because you are subscribed to the Google Groups "CCADB Public" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/a/ccadb.org/d/msgid/public/DS0PR14MB62166FC8869582923524D3EA9243A%40DS0PR14MB6216.namprd14.prod.outlook.com.
smime.p7s
Description: S/MIME cryptographic signature
