Re: [TLS] I-D Action: draft-ietf-tls-chacha20-poly1305-00.txt

[[e-mail] (Martin Rex)]

Hubert Kario wrote:

>>> 
>>> That's a violation of the engineering requirements :) If you don't
>>> need server authenticity, then you don't ask for it. If you ask for
>>> it, then you have to validate it.
>> 
>> It may appear that way -- but implementing support for ANON ciphers is
>> more work than NOT implementing support for certificate validation.
>> Supporting ANON in standards, implementation, testing/QA etc carry a
>> cost.  We should strive towards offsetting costs with use-cases that can
>> motivate the costs.
> 
> From QA perspective it's easier to test a specific ciphersuite than to test 
> some obscure option that disables signature verification that every 
> application names differently.
> 
> A(EC)DH ciphersutes are well known and used configuration option, "disable 
> signature verification on ServerKeyExchange" is not.


There may be SSL/TLS implementations that do not implement any of the
DH_anon/ECDH_anon cipher suites.  Our implementation does not implement them.
So for use with SMTP+TLS (which we do support) DH_anon may simply not be an
option.

But we also do not allow application programmers to skip certificate
path validation (that is only possible at the low-level APIs and used
by the SMTP+TLS/STARTTLS code.

The problem that I have with DH_anon/ECDH_anon is that it is equal
to TLS with Alzheimer.  Not being able to authenticate the server
is a problem, but also being unable to re-recognize a previous
communication peer is significantly worse.


-Martin