Re: [TLS] TLS 1.3 certificate delegation?

[Andy Lutomirski <[e-mail]>]

On Fri, Nov 8, 2013 at 2:40 PM, Rob Stradling <[e-mail]> wrote:
> On 08/11/13 17:43, Andy Lutomirski wrote:
> <snip>
>
>> The problem AIUI with X.509 constraints in general is that they only
>> apply to the next certificate down the chain,
>
>
> Disagree.  I'd say the reverse is true.
>
> The Basic Constraints pathLenConstraint field (RFC5280 4.2.1.9)...
>   "...gives the maximum number of non-self-issued intermediate
>    certificates that may follow this certificate in a valid
>    certification path."
>
> The Name Constraints extension (RFC5280 4.2.1.10)...
>  "...indicates a name space within which all subject names in
>   subsequent certificates in a certification path MUST be located."
>
>
>> which means that there's
>> plenty of room to get the spec wrong and make it useless [1].
>
>
> OpenSSL 0.9.8 doesn't process Name Constraints, but OpenSSL 1.0.0+ does.  So
> upgrade.  :-)
>
>
>> If TLS 1.3 adds a new mechanism here,
>
>
> ...then implementers might "get the spec wrong and make it useless"?

If it's widely used (possibly even by default in some implementation)
to use short-term fixed DH parameters a la QUIC, then implementations
won't have the opportunity to screw it up.

--Andy