Re: [pkix] reminder

[Scott Schmit <[e-mail]>]

On Sat, Aug 04, 2012 at 08:41:34AM -0700, Paul Hoffman wrote:
> On Aug 3, 2012, at 6:16 PM, Scott Schmit wrote:
> > On Thu, Aug 02, 2012 at 01:06:43PM -0400, Stephen Kent wrote:
> >> Add the following paragraph to the end of RFC 5280, Section 3.2:
> >> 
> >> Consistent with Section 3.4.61 of X.509 (11/2008) we note that use
> >> of self-issued certificates and self-signed certificates issued by
> >> other than CAs are outside the scope of this specification. Thus,
> >> for example, a web server or client might generate a self-signed
> >> certificate to identify itself. These certificates, and how a
> >> relying party uses them to authenticate asserted identities, are
> >> both outside the scope of RFC 5280.
> > 
> > As I read it, if you aren't a CA and generate a self-signed certificate,
> > then that certificate isn't a PKIX certificate.  Is that statement
> > right?
> 
> Partially. Those certificates are outside the scope of PKIX. However...
> 
> > If so, it sort of screws up draft-ietf-dane-protocol:
> > 
> > From section 2.2.1:
> > #  The certificate usages defined in this document explicitly only apply
> > #  to PKIX-formatted certificates in DER encoding [X.690].  If TLS
> > #  allows other formats later, or if extensions to this RRtype are made
> > #  that accept other formats for certificates, those certificates will
> > #  need their own certificate usage values.
> > 
> > (TLSA Certificate usage 3 is intended for self-signed "PKIX"
> > certificates.)
> 
> No. It is intended for PKIX-formatted certificates, which is exactly
> what the DANE spec says.
> 
> > Maybe the problem is that we talk about PKIX-formatted certificates vs.
> > PKIX-validated certificates, but both are handled in RFC 5280.
> 
> Can you say why that's a problem?

Because someone could easily use that paragraph to say, "there's no such
thing as a PKIX-formatted certificate that isn't PKIX-validatable." 

RFC 5280 == PKIX, so being out of scope of RFC 5280 makes it not
PKIX-anything.  After all, how the certificate is signed determines if
it's in scope of PKIX.

> > Can we distinguish the two?
> 
> Yes, easily.

I'd suggest we do that.  I think think that's why people were trying to
inject "use of" the certificates instead of "generation of" the certs--
to make that distinction.

I'd be happier if the last sentence read:

"Validation of these certificates, and how a relying party authenticates
the asserted identities, are both outside the scope of RFC 5280."

Then it makes sense for a certificate to be a PKIX[-formatted]
certificate, but not PKIX-validatable.

-- 
Scott Schmit