[Servercert-wg] Discussion Period Begins: Ballot SC-061v3: New CRL Entries must have a Revocation Reason Code

Wed Feb 1 21:55:59 UTC 2023

I agree with Wendy’s assertion that superseded isn’t the most appropriate reasonCode for “The Certificate no longer complies with the requirements of [Section 6.1.5](#615-key-sizes) and [Section 6.1.6](#616-public-key-parameters-generation-and-quality-checking) (CRLReason #4, superseded)”. If we’re doing something drastic as, say, banning RSA-2048 that most certainly means that those keys are now suspect and can no longer be trusted. Although the “weakAlgorithm”/”weEkalgorithm” reasonCode in later versions of X.509 would be more appropriate, I think that “keyCompromise” would be the most appropriate code defined in RFC 5280/X.509 2005-08.

Thanks,

Corey

From: Servercert-wg <servercert-wg-bounces at cabforum.org> On Behalf Of Aaron Gable via Servercert-wg
Sent: Wednesday, February 1, 2023 3:02 PM
To: Tim Hollebeek <tim.hollebeek at digicert.com>
Cc: CA/B Forum Server Certificate WG Public Discussion List <servercert-wg at cabforum.org>
Subject: Re: [Servercert-wg] Discussion Period Begins: Ballot SC-061v3: New CRL Entries must have a Revocation Reason Code

For better or worse, RFC 5280 says nothing about the meanings of the various revocation reasons, other than their names. X.509 itself (https://www.itu.int/rec/T-REC-X.509-201910-I/en <https://avanan.url-protection.com/v1/url?o=https%3A//www.itu.int/rec/T-REC-X.509-201910-I/en&g=ODI2NDBiZDYyNTJjMDQ5YQ==&h=OTJhN2RiNGNhYmI1MjI4NTYzMDVmNGRlMzY4MmZhZDNlMzEyMzI2ZTQwNjQ3MTA0YmZiNGY2ZjA3ZjM2Y2ZhYg==&p=YXAzOmRpZ2ljZXJ0OmE6bzo1Mjk0YmM0NGRlZGRhNzY2ODJlMTEyMDczNDkzYmE4Yjp2MTpoOkY=> ) has this to say:

The following reason code values indicate why a public-key certificate was revoked:
– unspecified can be used to revoke public-key certificates for reasons other than the specific codes.
– keyCompromise is used in revoking an end-entity public-key certificate; it indicates that it is known or
suspected that the subject's private key, or other aspects of the subject validated in the public-key
certificate, have been compromised.
– cACompromise is used in revoking a CA certificate; it indicates that it is known or suspected that the
subject's private key, or other aspects of the subject validated in the CA certificate, have been
compromised.
– affiliationChanged indicates that the subject's name or other information in the public-key certificate
has been modified but there is no cause to suspect that the private key has been compromised.
– superseded indicates that the public-key certificate has been superseded but there is no cause to suspect
that the private key has been compromised.
– cessationOfOperation indicates that the public-key certificate is no longer needed for the purpose for
which it was issued but there is no cause to suspect that the private key has been compromised.
– privilegeWithdrawn indicates that a public-key certificate was revoked because a privilege contained
within that public-key certificate has been withdrawn.
– aACompromise is only relevant for ACRL entries (see 17.2.3.1).
– weekAlgorithm indicates that the public-key certificate was revoked due to a weak cryptographic
algorithm and/or key (e.g., due to short key length or unsafe key generation).

Aaron 

On Wed, Feb 1, 2023 at 11:36 AM Tim Hollebeek <tim.hollebeek at digicert.com <mailto:tim.hollebeek at digicert.com> > wrote:

Well, if Mozilla policy is wrong, we shouldn’t do the wrong thing just because it’s already in Mozilla policy.  It just needs we would need a bit of coordination with Mozilla to fix it.  However I don’t believe that’s necessary.

If you read the highlighted section Wendy posted, it’s pretty clear that RFC 5280 explicitly intends for superseded to be (mis)used in this manner.  Only the first unhighlighted part before the first “or” requires a request for a new certificate.

I share Aaron’s concern about the fact that this code means three things, two of them unrelated to the other, but unfortunately there doesn’t appear to be an IANA expansion point here to allow easy allocation of a new reason code.  However a short RFC could allocate a “compliance (9)” reason code and describe its proper use, if desired.

I hope I don’t have to state that the CABForum unilaterally choosing to use an undocumented reason code would be highly undesirable.

-Tim

From: Servercert-wg <servercert-wg-bounces at cabforum.org <mailto:servercert-wg-bounces at cabforum.org> > On Behalf Of Aaron Gable via Servercert-wg
Sent: Wednesday, February 1, 2023 1:30 PM
To: Ben Wilson <bwilson at mozilla.com <mailto:bwilson at mozilla.com> >; CA/B Forum Server Certificate WG Public Discussion List <servercert-wg at cabforum.org <mailto:servercert-wg at cabforum.org> >
Subject: Re: [Servercert-wg] Discussion Period Begins: Ballot SC-061v3: New CRL Entries must have a Revocation Reason Code

I will admit that I also don't *love* using Superseded for this purpose. However:

- There is no other revocation reason that is better: it's certainly not keyCompromise, affiliationChanged, cessationOfOperation, or privilegeWithdrawn. Using "unspecified" for such incidents also feels bad, as this lumps together CA-initiated revocations due to incidents with the majority of Subscriber-requested revocations.

- This ballot is attempting to incorporate current Mozilla requirements into the BRs. The MRSP currently says that Superseded should be used for this purpose. If the BRs choose to diverge from MRSP in this way, there will be contradictory requirements until one or the other changes.

- One could choose to interpret Superseded as, rather than a declaration that the CA already has replaced the certificate, a declaration that they're *willing* to. A no-fault revocation, for lack of a better term.

None of these are particularly strong arguments in favor of using Superseded for this purpose, but together I think they make it the most reasonable option.

Aaron

On Wed, Feb 1, 2023 at 10:01 AM Ben Wilson via Servercert-wg <servercert-wg at cabforum.org <mailto:servercert-wg at cabforum.org> > wrote:

Thanks, Wendy.

Do others want to chime in on these points?

Ben

On Wed, Feb 1, 2023, 10:45 AM Wendy Brown - QT3LB-C <wendy.brown at gsa.gov <mailto:wendy.brown at gsa.gov> > wrote:

Superseded for these 2 reasons doesn't seem appropriate unless you also add that a new certificate was issued or at least requested, as a replacement.

6. The Certificate no longer complies with the requirements of [Section 6.1.5](#615-key-sizes) and [Section 6.1.6](#616-public-key-parameters-generation-and-quality-checking) (CRLReason #4, superseded);

and

12. The CA is made aware that the Certificate was not issued in accordance with these Requirements or the CA's Certificate Policy or Certification Practice Statement (CRLReason #4, superseded);

The definition isn't clear that a new cert has been issued (or even requested) based on the highlighted text - was it supposed to say that the CA issued a replacement certificate because it has reasonable evidence ....? Or should it just have ended with a . before the ", or the CA has ..."

  * **superseded (RFC 5280 CRLReason #4):** Indicates that the Certificate Subscriber has requested a new Certificate to replace an existing Certificate, or the CA has reasonable evidence that the validation of domain authorization or control for any fully‐qualified domain name or IP address in the Certificate should not be relied upon or the CA has revoked the Certificate for compliance reasons such as the Certificate does not comply with these Baseline Requirements or the CA's CP or CPS;

Thanks,

Wendy

Wendy Brown

Supporting GSA

FPKIMA Technical Liaison

Protiviti Government Services

703-965-2990 (cell)

On Wed, Feb 1, 2023 at 12:22 PM Aaron Gable via Servercert-wg <servercert-wg at cabforum.org <mailto:servercert-wg at cabforum.org> > wrote:

Wonderful, thank you! I have no further comments.

Aaron

On Tue, Jan 31, 2023 at 4:08 PM Ben Wilson <bwilson at mozilla.com <mailto:bwilson at mozilla.com> > wrote:

Thanks, Aaron - the numbering change was unintentional, so I fixed that, and I made other changes as requested.  See 

https://github.com/BenWilson-Mozilla/servercert/commit/f1ed2357c6c9fe9bcedaec040582f872e0f519de <https://avanan.url-protection.com/v1/url?o=https%3A//github.com/BenWilson-Mozilla/servercert/commit/f1ed2357c6c9fe9bcedaec040582f872e0f519de&g=YWViMDQzOWVjODgwNjI1Nw==&h=N2Q0ODVjYjQ2NGM1ZTM4NWM0MDQ5MTlmYmMyZWIzY2Q1YzM0YjhhMjBkNzdlOGMzNTE5YmMzMjlmZWM3MWFlNw==&p=YXAzOmRpZ2ljZXJ0OmE6bzpjODAzMjllZGMxNDZmMjFjOWU4MzU3NDA0ZDE4M2MwMzp2MTpoOkY=>  

Before I re-announce the discussion period, does anyone else have other changes that they would like to see?

Thanks, 

Ben

On Mon, Jan 30, 2023 at 9:58 AM Aaron Gable <aaron at letsencrypt.org <mailto:aaron at letsencrypt.org> > wrote:

The current redline appears to undo the recent renumbering of section 4.9.1.1, causing it to have two different instances of paragraphs 1 through 5. These were renumbered in Ballot SC-56 Cleanup[1]. Can we please preserve the new numbering?

Additional notes:

- In 4.1.1.1 (1), perhaps "without specifying a CRLReason", rather than "without giving a reason"? A Subscriber might state "Please revoke this because I accidentally deleted the keys", in which case they are giving a reason, but the best revocation reason is still 0 (Unspecified). One might believe that Superseded is applicable in this case, but that revocation request does not necessarily indicate that the Subscriber has also replaced the certificate.

- A very minor comment, but there's inconsistent phrasing between the five revocation reasons in Section 7.2.2: the first begins "Indicates that..." while the others begin "It is intended to be used...". Can we give all five of these entries the same structure/phrasing?

Aaron

[1] https://github.com/cabforum/servercert/pull/401/files#diff-e0ac1bd190515a4f2ec09139d395ef6a8c7e9e5b612957c1f5a2dea80c6a6cfeR1214-R1224 <https://avanan.url-protection.com/v1/url?o=https%3A//github.com/cabforum/servercert/pull/401/files%23diff-e0ac1bd190515a4f2ec09139d395ef6a8c7e9e5b612957c1f5a2dea80c6a6cfeR1214-R1224&g=M2UwYTBkMDgwN2Q5MTExYg==&h=MzhiZDYwZGUxNjhjYjlhZDgxOTkyNzVkMWRiZTUxODAzYmJkM2M4ZTc1MmZhMjkxYzBiZWJlYzE3YzQzY2NmYg==&p=YXAzOmRpZ2ljZXJ0OmE6bzpjODAzMjllZGMxNDZmMjFjOWU4MzU3NDA0ZDE4M2MwMzp2MTpoOkY=> 

On Thu, Jan 19, 2023 at 1:55 PM Ben Wilson via Servercert-wg <servercert-wg at cabforum.org <mailto:servercert-wg at cabforum.org> > wrote:

All,

This is version 3 of Ballot SC-061. I've moved some of the language down into section 7.2.2, and I've added back in two paragraphs that have been in the original Mozilla Root Store Policy regarding changing the reason code and revocation date for key compromise.  I also changed the compliance date to July 15, 2023. (The compliance date for CAs in Mozilla's program was Oct. 1, 2022.)

Purpose of Ballot SC-061 v.3

The purpose of this ballot is to modify sections 4.9.1.1 and 7.2.2 of the Baseline Requirements to incorporate the CRL reason codes that Mozilla has adopted in section 6.1.1 of the Mozilla Root Store Policy. 

Motion 

The following motion has been proposed by Ben Wilson of Mozilla and endorsed by David Kluge of Google Trust Services and Kiran Tummala of Microsoft.

—–Motion Begins—–

This ballot modifies sections 4.9.1.1 and 7.2.2 of the “Baseline Requirements for the Issuance and Management of Publicly-Trusted Certificates” as defined in the following redline, based on Version 1.8.6:

https://github.com/cabforum/servercert/compare/2c63814fa7f9f7c477c74a6bfbeb57e0fcc5dd5b..b1a3d9b491c9744a50a0e194678d76c639d6076b <https://avanan.url-protection.com/v1/url?o=https%3A//github.com/cabforum/servercert/compare/2c63814fa7f9f7c477c74a6bfbeb57e0fcc5dd5b..b1a3d9b491c9744a50a0e194678d76c639d6076b&g=ODQ0Mzc1NzQ3OGVhNjQyZg==&h=OGYwYjI0YmRmYjUwNzllMzU5MTkwOWQ4ZTZkYjEwMGVjN2IwNzg3ZDc5ZDA1ZDZkNjJlZjc0MzBkYjAzOTNlZA==&p=YXAzOmRpZ2ljZXJ0OmE6bzpjODAzMjllZGMxNDZmMjFjOWU4MzU3NDA0ZDE4M2MwMzp2MTpoOkY=> 

 —–Motion Ends—–

This ballot proposes a Final Maintenance Guideline. The procedure for approval of this ballot is as follows: 

Discussion (7+ days)

Start Time:  January 19, 2023 22:00 UTC

End Time: January 26, 2023 22:00 UTC

Vote for approval (7 days)

Start Time:  January 26, 2023 TBD

End Time: February 2, 2023 TBD

_______________________________________________
Servercert-wg mailing list
Servercert-wg at cabforum.org <mailto:Servercert-wg at cabforum.org> 
https://lists.cabforum.org/mailman/listinfo/servercert-wg <https://avanan.url-protection.com/v1/url?o=https%3A//lists.cabforum.org/mailman/listinfo/servercert-wg&g=ZjM2YTU2YmIyNDQwOWVhMA==&h=YWJmMDk1Mjc0OWU1MTM1ZmVlYmY3YTU1MTc0MzdjMzkxYTk4Mjc4NTVmNzE2YWRlZGE1Yzc3MTRmMmU3MjhkNw==&p=YXAzOmRpZ2ljZXJ0OmE6bzpjODAzMjllZGMxNDZmMjFjOWU4MzU3NDA0ZDE4M2MwMzp2MTpoOkY=> 

_______________________________________________
Servercert-wg mailing list
Servercert-wg at cabforum.org <mailto:Servercert-wg at cabforum.org> 
https://lists.cabforum.org/mailman/listinfo/servercert-wg <https://avanan.url-protection.com/v1/url?o=https%3A//lists.cabforum.org/mailman/listinfo/servercert-wg&g=NzhhMDFhZTQxMjk2YzRkNw==&h=NmFkMjZmNDIyNjhmZjg4MzI4ZjVlOTA4MzQ3ZGQ3NTIyYWUxZTliMTYyMDc4NmE2MzZjMTE4MTU1MjU5Mjg4NA==&p=YXAzOmRpZ2ljZXJ0OmE6bzpjODAzMjllZGMxNDZmMjFjOWU4MzU3NDA0ZDE4M2MwMzp2MTpoOkY=> 

_______________________________________________
Servercert-wg mailing list
Servercert-wg at cabforum.org <mailto:Servercert-wg at cabforum.org> 
https://lists.cabforum.org/mailman/listinfo/servercert-wg <https://avanan.url-protection.com/v1/url?o=https%3A//lists.cabforum.org/mailman/listinfo/servercert-wg&g=M2IyNjQyM2EzN2Q3YmUxNg==&h=NzcwNzMxNGExYzMzN2Y2NDY1MDdjMGZjOGUxMDRkZGM5NmQwYTIzNTFmNWRiYzMzMjc4YTM3ZDE2NzE2NWYxMQ==&p=YXAzOmRpZ2ljZXJ0OmE6bzpjODAzMjllZGMxNDZmMjFjOWU4MzU3NDA0ZDE4M2MwMzp2MTpoOkY=> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cabforum.org/pipermail/servercert-wg/attachments/20230201/2634dd7c/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4990 bytes
Desc: not available
URL: <http://lists.cabforum.org/pipermail/servercert-wg/attachments/20230201/2634dd7c/attachment-0001.p7s>