[TLS] Re: Adoption call for TLS 1.2 Update for Long-term Support

[Yaron Sheffer <[e-mail]>]

Hi Peter,

 

Just to put matters straight, the predecessor of RFC 9325, RFC 7525, was published in May 2015. But that doesn’t matter a whole lot now.

 

My point was much broader though: the IETF is sending deployers a bunch of mixed messages, and this is on us as a community.

 

RFC 9325 basically tells them: we prefer that you switch to TLS 1.3, but if you absolutely cannot do that, here’s how you can configure the existing TLS 1.2 and be secure (as of the time of publication).

 

TLS-LTS sends a whole different message of course.

 

And then the working group keeps nibbling at TLS 1.2 with documents like draft-ietf-tls-deprecate-obsolete-kex and the earlier “deprecating” documents. The KEX document does mention RFC 9325 at one point but does not say explicitly which of its requirements are new, making it hard for implementers to navigate our recommendations.

 

Thanks,

                Yaron

 

On 22/11/2024, 12:06, "Peter Gutmann" <[e-mail]> wrote:

Yaron Sheffer <[e-mail]> writes:

 

>Specifically, RFC 9325 [1] published a mere two years ago is not even

>referenced in the draft, let alone a comparison made with these deployment

>recommendations that were made by the very same IETF. (Yes you can hear my

>frustration coming through).

 

In defence of the -LTS draft, RFC 9325 postdates it by six years, so there

wasn't anything to reference at the time.  I'm also not certain how much

overlap there is between the two, for example 9325 contains quite a lot of

stuff (older TLS versions, compression, DTLS, fallback, RC4, NULL cipher

suites, RSA key transport, etc) that has no bearing on what's in -LTS which

means it could cause confusion if someone tries to apply it to things that

mostly don't exist in -LTS.

 

Having said that, now that my attention has been drawn to it :-), I'd be happy

to include a note along the lines of "further advice on secure use of TLS may

be found in RFC 9325", it would certainly fit in with what -LTS is trying to

achieve.

 

Peter.