[cabfpub] LV Certificates

[Eric Mill]

On Fri, Dec 18, 2015 at 5:42 PM, Peter Bowen <pzb at amzn.com> wrote:

> The proposers make the claim "anywhere from 2-7% of global user agents are
> unable to use HTTPS sites utilizing SHA-2 signature algorithms”.  It would
> be helpful to have more concrete data.  Would CloudFlare and Facebook be
> willing to share data on the top user agents (including versions and OS
> platform) which would be receiving these LV certificates, if they are
> approved?  Both have publicly stated they are using the type of algorithm
> proposed in the ballot, so such data should not be hard to get.
>

This sort of data will be crucial to evaluating the proposed ballot. The
level of data published on Facebook's and CloudFlare's original blog posts
isn't sufficient to do that.

In particular, we need to break down the different root causes of SHA-2
incompatibility, so that we can talk about how to most effectively mitigate
each one.

For example: some CAs are removing previously publicly trusted root
certificates from browser trust stores, apparently (at least in part) to
continue issuing SHA-1 certificates to customers who have clients whose
trust stores will continue to include those root certificates.

So, the segment of LV's target audience whose constraint is that they
aren't in a position to upgrade their browser or OS may be an excellent
candidate for certificates issued from these roots, since presumably their
trust store is also not receiving upgrades.

The finite number of ex-publicly-trusted roots that can offer this service
doesn't make this a very liquid market, but since we're trying to mitigate
a hopefully temporary situation, this seems like a better direction to go
than introducing a new certificate class into the global CA system.


Other clients who can't make SHA-2 connections may be behind intermediate
hardware/software endpoints, such as a network middlebox or an antivirus
program. These users may have updated trust stores and operate clients that
ostensibly support SHA-2, but are powerless to affect (or are unaware of)
the intermediate endpoint in their way.

If that's the case, we should examine the extent to which those clients are
part of the "disadvantaged, war-torn areas of the world" that the ballot's
preamble offers as its primary justification.

Given the CA/B Forum's previous withdrawal of a ballot that proposed a
generic extension of the SHA-1 issuance deadline, and the consensus that
appeared to exist that the general concerns of large enterprises over
compatibility were not sufficient to merit an extension of SHA-1 issuance,
it seems very important to establish how effective LV would be at serving
the preamble's stated target audience.

-- Eric


> > On Dec 18, 2015, at 2:36 PM, Jeremy Rowley <jeremy.rowley at digicert.com>
> wrote:
> >
> > For now I’m presenting it on behalf of non-members of the Forum, but I
> will likely sponsor if I get the necessary internal approvals.
> >
> > From: Ryan Sleevi [mailto:sleevi at google.com]
> > Sent: Friday, December 18, 2015 3:24 PM
> > To: Jeremy Rowley
> > Cc: CABFPub
> > Subject: Re: [cabfpub] LV Certificates
> >
> > Jeremy,
> >
> > Is this something DigiCert is endorsing, or are you merely presenting it
> on behalf of non-members of the Forum in the effort to find sponsors and
> endorsers?
> >
> > On Fri, Dec 18, 2015 at 2:21 PM, Jeremy Rowley <
> jeremy.rowley at digicert.com> wrote:
> > Hi everyone,
> >
> > Attached is a proposal from Cloudflare and Facebook creating LV
> certificates in the baseline requirements.  This is a draft ballot for
> review that will, of course, change based on the debate in the forum.
> Although CAs will stop issuing SHA-1 on 2016/1/1, there isn’t any reason
> these changes couldn’t go into effect in early January (assuming a passing
> vote).
> >
> > If adopted, this ballot would permit continued use of SHA1 certificates
> past the deprecation deadline (to support older devices) but give newer
> browsers an easy way to reject SHA1 for users.  The ballot also increases
> the resiliency of SHA1 certs against attacks by requiring higher entropy
> serial numbers.
> >
> > I look forward to your comments.
> >
> > Thanks,
> > Jeremy
> _______________________________________________
> Public mailing list
> Public at cabforum.org
> https://cabforum.org/mailman/listinfo/public
>



-- 
konklone.com | @konklone <https://twitter.com/konklone>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://cabforum.org/pipermail/public/attachments/20151219/a8274a5b/attachment-0001.html