Re: [TLS] consensus on backwards compatibility changes

[Dave Garrett <[e-mail]>]

On Tuesday, January 27, 2015 11:42:22 am Joseph Salowey wrote:
> [Joe] I think this makes sense.  I added to comments to the PR.  I propose
> to move the bit about the server accepting versions {3,x} to the same place
> and change the wording of the existing test to say:
> 
> "The client MUST set the version to {3, 1} for the initial ClientHello."

Language changed and {3,X} backward compatibility sentence added.

> > 5) Explicitly prohibit EXPORT ciphers and any others <100 bits.
> > (100 bit line is arbitrary; could be 112 if preferred)

> [Joe] draft-ietf-uta-tls-bcp-08 recommends 112 so we probably would match
> that.

No objection. Updated.

On Tuesday, January 27, 2015 01:49:04 pm Martin Rex wrote:
> > "The client MUST set the version to {3, 1} for the initial ClientHello."
>
> The primary reason why several implementors goofed the TLS version
> negotiation in TLSv1.0 is that lack of clarity in the language.
> 
> If your above statement refers to the client_version member of the
> ClientHello handshake message, please say so.

The language is proposed for within the version field definition for the record
layer version.

Would we want a paragraph somewhere explicitly mentioning possible
confusion between record layer version, hello version, & negotiated version?

On Tuesday, January 27, 2015 02:20:38 pm Martin Rex wrote:
> >  Implementations MUST NOT send a ClientHello or ServerHello with
> >  the protocol version set to { 3, 0 } or less. Any endpoint receiving a
> >  Hello message with the protocol version set to { 3, 0 } MUST respond
> >  with a "protocol_version" alert message and close the connection.
> 
> A description similar to this caused the TLS version negotiation problems.
> The structure members in the handshake message PDUs are called
> ClientHello.client_version and ServerHello.server_version.
> 
> If the above paragraph is meant to apply to the structure members
> (client_version, server_version) of the respective TLS handshake message PDUs
> (ClientHello, ServerHello), then please use the names of these structure
> members for disambiguation.

Yes, Eric asked for that same change. I updated the PR a couple days ago.


Dave