[lamps] Re: TLS Client Certificates; a survey

[Tomas Gustavsson <[e-mail]>]

> We should admit that the CA infrastructure has failed us for nearly all use-cases.  Either the CA infrastructure is the web, and (despite the CA/B forum
> rules) it's OK to use web certs in non-web contexts.  Or, the CA infrastructure is more than the web, and we need to have new,, non-web CAs with rules
> outside of the CA/B forum.

There are tons of CAs outside of the WebPKI/CA/B Forum ecosystem. For web and non-web use cases. EU TSPs, X9 Financial PKI, Adobe, ICAO, just to mention a few well known. I think large parts of industry has in the last couple of years realized that Web PKI isn't the most robust choice for non-browser use cases and use-case specific alternatives are showing up more and more.
The PKI Consortium tried to establish a project for a "List of trust lists" a couple of years ago, but it didn't really get finished, but there is an archive here. https://github.com/pkic/ltl



________________________________
From: Nico Williams <[e-mail]>
Sent: Monday, March 23, 2026 11:09 AM
To: Michael Richardson <[e-mail]>
Cc: Salz, Rich <rsalz=[e-mail]>; Tls <[e-mail]>; [e-mail] <[e-mail]>
Subject: [lamps] Re: TLS Client Certificates; a survey

On Mon, Mar 23, 2026 at 02: 00: 35PM -0400, Michael Richardson wrote: > Salz, Rich <rsalz=40akamai. com@ dmarc. ietf. org> wrote: > > Since WebPKI CA’s will not be able to issue TLS-Client certificates, > > what are the customers


On Mon, Mar 23, 2026 at 02:00:35PM -0400, Michael Richardson wrote:
> Salz, Rich <rsalz=[e-mail]> wrote:
>       > Since WebPKI CA’s will not be able to issue TLS-Client certificates,
>       > what are the customers and CAs thinking of doing?
>
> You say this as if it's a new thing :-)

well, it's recent.  It happened around October 2025.

> Is it the "change" that certificates obtained for code signing or email use
> will have the tls-kp-clientAuth EKU ommitted?

The change is that roots in the Chrome Root Program may not sign
intermediates that sign any EE certificates having any EKUs other than
id-kp-serverAuth.  A CA can still have roots outside the CRP that do
sign intermediates that do sign non-server EE certificates, but probably
few will.

>     > Replies to be will be summarized to both lists. Please be careful if
>     > you use reply-all.
>
> 1. This assumes the RP are checking EKU.

Yes, but they should.

> 2. I think 94% of usage of mTLS is via private PKI for the client side.

Probably true.  The two applications I know to be affected are XMPP and
SMTP.

Nico
--

_______________________________________________
Spasm mailing list -- [e-mail]
To unsubscribe send an email to [e-mail]