Sorry, getting to this late. 

Thank you Jim for the feedback.

I think an good point comes out of this discussion. We need to articulate the 
RA role that could at the same time act as a proxy. An RA usually performs 
authentication/authorization before relaying to the CA that already has a 
predetermined relationship with. A proxy acts as a plain COAP2HTTP translator 
that does not necessarily perform any authentication or authorization of the 
client. In either case ESP PoP cannot be performed as DTLS to the client is 
terminated in the middle. 

In summary, reading this discussion I see these things to improve in the draft 
- Clarify DTLS 1.2 and 1.3 in the DTLS section.
- Clarify the use of COSE or CMS for server-side keys and chose the MTI. I 
don't think COSE just for the keys while we use ASN.1 DER for ca/enrolled certs 
will buy us much. And it will alter server-side key generation messages defined 
in RFC7030. I think introducing COSE for these messages is fine as long as it 
is conveyed in the response headers and they key establishment is clear. Given 
that we have a way of doing it based on RFC7030, I think this falls out of 
scope for this draft. 
- Add text about short and long names
- Proxy and RA clarification in Section 6.
- Be explicit about the proxy getting the whole message before forwarding in 
Section 6. 


-----Original Message-----
From: Ace [] On Behalf Of peter van der Stok
Sent: Thursday, March 15, 2018 6:11 AM
To: Michael Richardson <>
Subject: Re: [Ace] draft-ietf-ace-coap-est-00

Michael Richardson schreef op 2018-03-15 09:00:
> peter van der Stok <> wrote:
>     >> >> DTLS connection is going to be required to act as an RA.  RAs
>     >> are required
>     >> >> to have the entire request for adding authentication as 
> necessary.
>     >>
>     >> > This is visible in the figure of section 6, but needs 
> elaboration in
>     >> the
>     >> > text.
>     >>
>     >> I don't understand why we have that paragraph.
>     >> An end point that terminates the Pledge (D)TLS connection and 
> acts as
>     >> an RA *IS* a Join Registrar, not a Proxy.
>     >>
>     > Thus is outside the BRSKI context, and thus a proxy with RA 
> (separate or not)
> Let me delete "Join" from above sentence.
> A device that terminates the DTLS security (CoAPS) and then talks to 
> the CA is a Registration Authority according to EST and RFC5280.  It's 
> not a proxy.
> (And it doesn't matter if it speaks HTTPS or CMS or CMP or 
> super-pigeon-telepathy to the CA)
A http/coap proxy  is specified in RFC8075. It explains "how an HTTP request is 
mapped to
    a CoAP request and how a CoAP response is mapped back to an HTTP

In the est-coap draft DTLS and TLS connections are terminated in the http/coap 
proxy, and the proxy is therefore connected to an RA (possibly running on the 
same host as the proxy).

Where is my terminology going astray?


Ace mailing list

Ace mailing list

Reply via email to