Renewal is indeed important for EST-obtained certificates, and knowing when to renew also. In draft-ietf-anima-constrained-voucher (cBRSKI) we call it "re-enrollment" (this includes both plain renewal as well as a potential change of domain TA). Do we have this term correct?

If renewal happens frequently enough, we might have a good use case for using Uri-Path-Abbrev. But, this is only in the following case: 1. Device knows, or somehow discovers, address/port of the EST Registrar and contacts it using a /.well-known/est/... resource.

There's also other cases enabled by RFC 9148: (Section 4.1)
2. Device knows, or somehow discovers, address/port of the EST Registrar, makes DTLS connection, and uses CoAPS-secured discovery to get resource names.  (Shorter Link Format document - example 1 in 4.1. Still more verbose than really needed...) 3. Device uses unsecured CoAP discovery to find address/port/resources of EST Registrar, then makes DTLS connection and uses the discovered resource names. (Very long Link Format document - example 2 in 4.1.)

In case 2/3, Uri-Path-Abbrev won't be used. Not a problem, just wanted to point this out.

Esko

On 27-9-2025 19:32, Michael Richardson wrote:
Christian Amsüss<[email protected]> wrote:
     > I think this is as compact as it gets, with all the options of
     > repetition deferred to future development (sketched in an appendix for
     > the benefit of future documents), it's really just a simple number.

Agreed.

Rifaat,MikeO and I have just posted 
draft-yusef-lamps-rfc7030-renewal-recommendation.
It replicates RFC9773 (an ACME thing) into EST/RFC7030.

It (renewal-info) probably should also update RFC9148 to add a new renewal-info 
for
EST-coaps.   So this would fit into the path-abbrev 30x space!
I think it will be fine for renewal-info to block on publication of abbrev.
Probably renewal-info should return CBOR for 9148 use, so we'll need to do CDDL.

So I'll include IANA Considerations in that document.

----

I think that renewal-info will have to resurrect the well-known/EST registry
that RFC8995 created as I-D, then walked back to make well-known/BRSKI
instead near the last minute...

--
Michael Richardson<[email protected]>   . o O ( IPv6 IøT consulting )
            Sandelman Software Works Inc, Ottawa and Worldwide





_______________________________________________
core mailing list [email protected]
To unsubscribe send an email [email protected]

--
*IoTconsultancy.nl* | Email/Teams: [email protected] | +31 6 2385 8339
_______________________________________________
Ace mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to