chrysn

Thanks for the report and sorry I missed it initially. The non-builtin option access will generally need some love; they're probably quite tricky to use right now. Did you come...

Some input on this is captured in [a RIOT forum thread](https://forum.riot-os.org/t/coap-remote-shell/3340/9), especially from @mcr.

Right now there is no setting for that -- but then again, the RD itself has nothing to log, and all that'd show up is CoAP messages in the aiocoap...

Leaving this open as it is a good suggestion to improve the aiocoap-rd too.

Because registration updates per POST have no payload as per [registration update description](https://datatracker.ietf.org/doc/html/draft-ietf-core-resource-directory-28#section-5.3.1). There is currently no PATCH format specified for links (the section on [further operations](https://datatracker.ietf.org/doc/html/draft-ietf-core-resource-directory-28#section-5.3.3) has some more...

I just didn't get to it over the holiday backlog yet. I thin I get the issue, trying to come up with sth over the weekend.

> What is current error reporting mechanism of aiocoap library to report > to client (implemented using this library), that COAP server was > restarted, existing DTLS session lost and...

I've fixed a few missing pieces in the DTLS error handling, please try whether that solves your immediate issue. Of the three error paths I've observed with the libcoap example...

On the topic of reconnecting and where to handle errors: A network error *does* need to be handled by the application, simply because aiocoap can't (generally) be certain that "just...

Just to make sure as I see a few "connection refused" and "no credentials" errors: * With the same setup but aiocoap of this morning (say, version e181700a93a2f8ee5778cc3c8022861cc9454aba), do things...