Maybe this is already solved during the IETF week.
If not, read my answer below ..
Ladislav Lhotka <lhotka@xxxxxx> wrote:
> On Tue, 2019-11-19 at 13:26 +0800, Michael Richardson wrote:
>> Tim Wicinski <tjw.ietf@xxxxxxxxx> wrote: >> IANA has been doing it for
>> more than five years for a couple of YANG >> modules (six by now), and
>> I am not aware of any troubles so far.
>>
>> > But that's work put onto the IANA pile, and some I* organization may
>> > want to understand the commitment.
>>
>> And, I don't understand how the products get updated. Are they going
>> to pull YANG modules from IANA? How often: hourly? daily? yearly?
> Products get updated ad hoc, as soon as their implementors feel the
> need. This usually has to be accompanied with changes in the underlying
> system - actually implementing support for new registry entries etc.
So if software (including management software) needs explicit support for new
IANA entries (in for instance, lists of ciphers), then why is it better for
the (human) programmer to download a new YANG module than to just read the IANA web
pages (or .XML equivalent)?
The point is that you download the IANA-maintained YANG modules on
regular basis (ex: typical interface types) for the management software
that speaks YANG. A cronjob does the job.
Since the new YANG module is backward compatible, the software works
automatically (for the new interface type) with the new YANG module.
Exactly like the IfType defintions for MIB modules (see
https://www.iana.org/assignments/smi-numbers/smi-numbers.xhtml).
Regards, B.
(It's not like the YANG description will have executable code for SHA4 or something)
