Re: [TLS] Verifying X.509 Certificate Chains out of order

[Eric Rescorla <ekr at networkresonance.com>]

At Tue, 7 Oct 2008 13:43:53 +0200 (MEST),
Martin Rex wrote:
> 
> Martin Rex wrote:
> > 
> > Stefan Santesson wrote:
> > > 
> > > Just agreeing on the principle that implementers should be forced to
> > > send the certificates in order but it definitely must be allowed
> > > to accept out of order chains.
> > 
> > I have absolutely no problem with implementations that accept an
> > unordered list.
> 
> Thinking about it, what exactly do you mean with unordered?
> 
> Since there isn't any additional information in the protocol to
> identity the end-entity cert in the certificate_list, that certificate
> will have to be the first.  Or does your code really apply heuristics
> in locating the end entity cert?

This is an excellent point. It does seem rather problematic to
have multiple candidate EE certs, especially because in many
cases the SSL stack does not know the peer's expected name.

-Ekr
_______________________________________________
TLS mailing list
TLS at ietf.org
https://www.ietf.org/mailman/listinfo/tls