I've now updated the cBRSKI draft to require (MUST) Uri-Path-Abbrev
support on the Registrar, such that a Pledge that wants to use it, can
use it without causing failures or additional round trips (which would
create a problem worse than the problem this option is trying to solve).
The new text is here:
https://datatracker.ietf.org/doc/html/draft-ietf-anima-constrained-voucher#section-6.3.3
For a non-constrained host such as the Registrar this is fine and it has
benefits for all cBRSKI/EST-coaps messages. The EST-coaps interactions
also occur during the lifetime of the device, after onboarding, e.g. for
certificate renewal.
What I've now remarked for the cBRSKI Join Proxy issue 86
<https://github.com/anima-wg/constrained-join-proxy/issues/86> is that I
don't see it as useful there to require support for this feature for the
CoAP discovery operation.
The discovery request is still one short 6LoWPAN frame, with or without
the path abbreviation.
Also, specific profiles of cBRSKI might use their custom way to discover
the Join Proxy. (E.g. in Thread mesh networks, it's something
Thread-specific not using CoAP.) So there's less incentive trying to
highly optimize this leg.
Esko
On 3/4/26 17:54, Michael Richardson wrote:
Esko, do you think that there is still a reason to do Uri-Path-Abbrev at all?
Esko Dijk<[email protected]> wrote:
> Issue has been presented for discussion at IETF 124, Montreal, in
> slides:
>https://datatracker.ietf.org/meeting/124/materials/slides-124-anima-01-constrained-brski-cbrski-and-join-proxy-00
> It's remarked there that the size savings for the request message is
modest. Besides,
> * the request only travels over a single hop, so there is no accumulated
savings over multiple hops.
> * the request fits in a single 6LoWPAN frame anyhow - there's no
potential reduction in 6LoWPAN fragments by applying the option.
> * the savings in power and reduced RF channel use (for a typical 250
kbps 802.15.4 6LoWAN) for an action only used during onboarding time, are very
minimal over the entire device lifetime.
> * it would require a constrained device to support the Option -- a JP,
as opposed to a Registrar as for the case of cBRSKI where this Option is applied
see [link](https://github.com/anima-wg/constrained-voucher/issues/336).
> * a particular profile of cBRSKI, that defines its own discovery
methods, can still introduce the option if they need it.
> Given all these considerations, it's proposed to *not* require the
Uri-Path-Abbrev option for discovery. And not make a text update.
> --
> Reply to this email directly or view it on GitHub:
>https://github.com/anima-wg/constrained-join-proxy/issues/86#issuecomment-3983167615
> You are receiving this because you were mentioned.
> Message
ID:<anima-wg/constrained-join-proxy/issues/86/[email protected]>
> ----------------------------------------------------
> Alternatives:
> ----------------------------------------------------
--
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]
_______________________________________________
Anima mailing list -- [email protected]
To unsubscribe send an email to [email protected]
