Re: [netmod] YANG library structure

[Ladislav Lhotka <[e-mail]>]

Martin Bjorklund <[e-mail]> writes:

> Hi,
>
> There has been quite a lot of discussion about the YANG library
> data model on the list.  The authors of draft-ietf-netconf-rfc7895bis
> have tried to understand all arguments in the discussion, and provide
> a solution.  Below are 3 solution proposals (we have discussed more,
> but they are basically just variations on the same themes).
>
> Absolute Requirements
> ---------------------
>
> o  RFC 7950, Section 5.6.5 says:
>
>      A server MUST NOT implement more than one revision of a module.

I believe YANG library should be liberated from the dependence on
servers and protocols and become part of a formal data model
specification. It is not just some state data but rather metadata, and
it is also perfectly conceivable that a server does not publish its YANG
library at all (e.g. for security reasons or constrained devices).

And then, I would suggest to consider dropping the above requirement
for the *general* YANG library, i.e. design it so that it can in
principle support multiple revisions of the same module. This is
important because

- it allows for supporting multiple revisions of some hardware
  (e.g. line cards) in the same device

- the server needn't represent just a single device but may be used for
  configuring a network of devices, and then the above restriction could
  be severely limiting.

This said, it would still be possible for a specific protocol to impose
such a restriction.

...

>
> Alt. B.
> -------
>
>   Each datastore refers to a schema, and each schema contains a list
>   of references to module-sets, and each module-set contains a flat
>   list of all modules, features, etc.

I like this one because different schemas will often need to reuse the
same module sets, and it will also be quite easy to add schema mount
specification to this.

Lada

-- 
Ladislav Lhotka
Head, CZ.NIC Labs
PGP Key ID: 0xB8F92B08A9F76C67