[Servercert-wg] Ballot proposal: require distributionPoint in sharded CRLs

Aaron Gable aaron at letsencrypt.org
Mon Oct 17 19:53:12 UTC 2022


It looks like we have two additional endorsers (Corey Bonnell and Dimitris
Zacharopoulos). I've made two small changes to the GitHub PR:
* changed effective date to Jan 15 for a little breathing room around the
holidays, at Tim Hollebeek's suggestion
* added a reminder that the IDP extension must be marked critical, at Wayne
Thayer's suggestion.

Since it looks like this will not result in any merge conflicts with the
profiles ballot, are there any other comments before we assign this a
number and open the official ballot?

Thanks,
Aaron

On Fri, Oct 14, 2022 at 1:16 PM Corey Bonnell <Corey.Bonnell at digicert.com>
wrote:

> > Seems there may still be a residual small risk if a new CRL shard is
> started before CCADB is updated any certificate that should be covered in
> that new shard gets revoked prior to the CCADB update.  But maybe people
> perceive that timing issue as too small to be a concern.
>
>
>
> I agree this is a risk and I think this is an area that should be
> clarified, especially since the CCADB and Root Program policies are silent
> on the order in which new sharded CRLs can be populated with entries and
> when those new shards need to be disclosed, or the relevant time frame in
> which such updates must occur. Hopefully CAs are updating CCADB with new
> CRL shard URIs prior to populating them with entries but clarifying this
> would eliminate any risk from certificate revocations not being processed
> due to those shards not being timely disclosed in CCADB.
>
>
>
> Thanks,
>
> Corey
>
>
>
> *From:* Servercert-wg <servercert-wg-bounces at cabforum.org> *On Behalf Of *Wendy
> Brown - QT3LB-C via Servercert-wg
> *Sent:* Friday, October 14, 2022 3:58 PM
> *To:* Aaron Gable <aaron at letsencrypt.org>
> *Cc:* CA/B Forum Server Certificate WG Public Discussion List <
> servercert-wg at cabforum.org>
> *Subject:* Re: [Servercert-wg] Ballot proposal: require distributionPoint
> in sharded CRLs
>
>
>
> So just to be clear, this change proposal is only trying to address the
> ability of a browser relying on CRL disclosures in CCADB to be able to
> ensure they have the complete set of CRLs disclosed there, not to address
> the potential risk to any given revoked certificate not being seen as
> revoked because the RP is looking at a CRL that does not have that
> certificate in scope due to sharding?
>
>
>
> Seems there may still be a residual small risk if a new CRL shard is
> started before CCADB is updated any certificate that should be covered in
> that new shard gets revoked prior to the CCADB update.  But maybe people
> perceive that timing issue as too small to be a concern.
>
>
>
> thanks,
>
> Wendy
>
>
>
> Wendy Brown
>
> Supporting GSA
>
> FPKIMA Technical Liaison
>
> Protiviti Government Services
>
> 703-965-2990 (cell)
>
>
>
>
>
> On Fri, Oct 14, 2022 at 3:48 PM Aaron Gable <aaron at letsencrypt.org> wrote:
>
> On Fri, Oct 14, 2022 at 12:34 PM Wendy Brown - QT3LB-C <
> wendy.brown at gsa.gov> wrote:
>
> Just a question -
>
> if a certificate that is being checked for revocation does not contain a
> cDP, how will requiring iDP in the CRL assist in preventing a CRL
> substitution attack? If you don't have the correct cDP for a given
> certificate how will the iDP in that sharded CRL provide assurance that the
> RP is looking at the correct CRL?
>
>
>
> In the case of the CRLs disclosed in CCADB's JSON Array of Partitioned
> CRLs field, the relying party (e.g. Mozilla or Apple) can verify that the
> distributionPoint contained within the CRL matches the URL disclosed in
> CCADB.
>
>
>
> On Fri, Oct 14, 2022 at 11:14 AM Corey Bonnell <Corey.Bonnell at digicert.com>
> wrote:
>
> I don’t believe the profiles ballot modifies section 7.2 at all, so there
> should be no conflict in having a separate proposal.
>
>
>
> The current profiles ballot lightly modifies Section 7.2.1 (
> https://github.com/cabforum/servercert/pull/373/files#diff-e0ac1bd190515a4f2ec09139d395ef6a8c7e9e5b612957c1f5a2dea80c6a6cfeR3118),
> but not in a way that would lead to a merge conflict with this ballot.
>
>
>
> Aaron
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cabforum.org/pipermail/servercert-wg/attachments/20221017/d087745c/attachment-0001.html>


More information about the Servercert-wg mailing list