Re: [TLS] Comments/Questions on draft-gutmann-tls-encrypt-then-mac-00.txt

[[e-mail] (Martin Rex)]

Michael D'Errico wrote:
> > http://www.ietf.org/internet-drafts/draft-bmoeller-tls-downgrade-scsv-00.txt
> 
> I think that the server MUST send a FATAL alert only if it would
> not have been willing to negotiate the lower TLS version in the
> absence of the SCSV.
> 
> A WARNING alert from the server (or some extension_data with more
> information) lets the client decide whether to continue.  Both
> sides can keep track of these occurrences for further investigation
> by interested admins at their leisure (not via calls to the help
> desk).
> 
> Also RFC 3546 and 4366 have been obsoleted by RFC 6066.

"Warning level alerts"?  Please don't!  As rfc6066 summarizes,
the last experiment of using warning level alerts failed badly
  -> "unpredictable client behaviour" and it is NOT RECOMMENDED.


Speaking of rfc 3546/4366/6066 "TLS Extensions":

  https://tools.ietf.org/html/rfc6066#section-3

                  If the server understood the ClientHello extension but
   does not recognize the server name, the server SHOULD take one of two
   actions: either abort the handshake by sending a fatal-level
   unrecognized_name(112) alert or continue the handshake.  It is NOT
   RECOMMENDED to send a warning-level unrecognized_name(112) alert,
   because the client's behavior in response to warning-level alerts is
   unpredictable. 

-Martin