On Mon, 10 Nov 2025 at 16:47, Michael Richardson <[email protected]> wrote:

>
> Heikki Vatiainen <[email protected]> wrote:
>


>     > certificate is not used. Should this variant of EAP-TLS have its own
>     > EAP type?
>
> As we discussed at the mic, I'm not convinced it's needed, and I'm
> concerned
> that it requires much more effort to deploy.
>

After reading the other messages in this email thread, I no longer think
it's a good idea to have a new EAP type:
- it requires more to deploy, as you've mentioned above, and as Alan wrote
a new EAP type  would require more code than using an existing type
- if e.g., TEAP were to have server-unauthenticated mode, would it need a
new EAP type? I don't think so
- EAP-FAST has server-unauthenticated mode and it has worked just fine with
all modes using the same EAP type (EAP-FAST)
- enabling client-unauthenticated and/or server-unauthenticated mode for an
EAP type is simply a configuration option with possible signalling with
eap.arpa identity or other means


>     > The Wi-Fi Alliance and hostapd specifications and implementations,
>     > discussed earlier on the list, appear to use an EAP type that is
>     > different from EAP-TLS type.
>
> Which method is this?
>

I'd say these two hostap git links show best what the currently available
server-unauthenticated EAP-TLS variants are:

1. UNAUTH-TLS vendor specific EAP type - this method includes only server
authentication
https://git.w1.fi/cgit/hostap/commit/?id=065d2895b4693e8c923580dbfa31123297c8bb7d

2. HS 2.0R2: Add WFA server-only EAP-TLS peer method
I think this is the method that is referenced by
https://www.ietf.org/archive/id/draft-ietf-emu-eap-arpa-10.html#section-1.1.1-3
https://w1.fi/cgit/hostap/commit/?id=8e5fdfabf69a7692d1a0d04f00fa10
3e9ff72010

-- 
Heikki Vatiainen
[email protected]
_______________________________________________
Emu mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to