Re: [TLS] External PSK importers

Happy to talk about this.

Based on a quick reading, a few notes:

You should do something more concrete with the label parameter.  Keep
in mind that both client and server need to agree on a use for this,
so my initial intuition to put the server identity might not work, but
it could be a start.  The problem being that how the client identifies
the server might not be something it shares with the server.

You equivocate regarding the hash to use.  If we're going to make this
solid, then we need to follow good hygiene and say that each imported
PSK needs to be used with just one hash function.

You equivocate again more than I like about TLS 1.2.  How about you
reserve a label specifically for TLS 1.2 and feed the PSK through the
expand function using that label?
On Tue, Oct 30, 2018 at 3:41 PM Christopher Wood <cawood@apple.com> wrote:
>
> To resurrect the discussion around external PSKs and TLS 1.3, I wrote down the importer-based design Ekr and I discussed [1].
>
>    https://datatracker.ietf.org/doc/draft-wood-tls-external-psk-importer/
>
> Time permitting, I’d like to present this in BKK. It would be useful to compare and contrast this against David’s Universal PSK draft [2].
>
> Thanks,
> Chris (chair hat off)
>
> [1] https://mailarchive.ietf.org/arch/msg/tls/FKB5oNxTVoIh59I3q9epHk6Q1Uw
> [2] https://datatracker.ietf.org/doc/draft-davidben-tls-universal-psk/
>
> _______________________________________________
> TLS mailing list
> TLS@ietf.org
> https://www.ietf.org/mailman/listinfo/tls