Max Pritikin (pritikin) <priti...@cisco.com> wrote:
> Section 3.1.1. Discovery Supports a backoff mechanisms but on review
> I’m thinking the language about final failure to be vague: "Once all
> discovered services are attempted the device SHOULD return to Multicast
> DNS. It should periodically retry the vendor specific mechanisms. The
> Pledge may prioritize selection order as appropriate for the
> anticipated environment."
Michael B and I had some long conversation about how manufacturers configure
new nodes. We are pretty sure that they will require either audit tokens,
or ownership vouchers, never both. As part of that definition, the
manufacturer would also decide what other vendor specific mechanisms will
exist.
So to emphasize your point, our role isn't to say what those mechanisms are,
(with the exception of mDNS, which I'd like to put into an appendix), but
rather when they get used in the state machine.
> Section 5.1 Request Voucher from the Registrar "As detailed in Section
> 3.1.1 if no suitable Registrar is found the Pledge restarts the state
> machine and tries again. So a Registrar that is unable to complete the
> transaction the first time will have future chances.”
> A flood telemetry status indicator in addition to direct feedback to
> the Registrar. This would enable any local equipment to better report
> the Pledge’s state. I suppose this SHOULD be signed but may be unsigned
> to avoid extra processing overhead on the Pledge. This would leak the
> Pledge’s identity information so there are privacy/security concerns
> but it does provide feedback.
I think that this can be decoupled from our work, and we could define this
later? I think it's pretty important, but I want to get it right.
> Normative text indicating physical indicators such as a blinking LED
> associated with each discovery state could be RECOMMENDED for capable
> devices. Obviously some devices simply won’t have such things so this
> can’t be required. Doing this would help clarify the discovery
> states. Although keep in mind the existing s3.1.1.1 recommendation that
> "Methods SHOULD be run in parallel to avoid head of queue problems”;
> meaning that the states indicated might be a generic “discovery”.
Glad you agree about the difficulties here, yet the opportunity is clear.
--
Michael Richardson <mcr+i...@sandelman.ca>, Sandelman Software Works
-= IPv6 IoT consulting =-
signature.asc
Description: PGP signature
_______________________________________________ Anima mailing list Anima@ietf.org https://www.ietf.org/mailman/listinfo/anima
