Re: [Rtg-yang-coord] comment concerning draft-shaikh-rtgwg-policy-model-00

[Anees Shaikh <[e-mail]>]

Right, those examples are relatively easy because they are defined by a
standard or widely used common settings. We try to do this in the
OpenConfig models but there are plenty of examples where every
implementation has chosen a different default, range, etc.

Thanks.
-- Anees
On Thu, May 28, 2015 at 1:18 PM Mahesh Jethanandani <[e-mail]>
wrote:

> Aneesh,
>
> When I look at a model to see where the ranges, defaults, min/max values
> can be defined, I rely on protocol definition and what the expectation is
> of the packet on the wire.
>
> Two examples come to mind. VLAN, as we know has a fixed range, and can be
> easily defined. Same for MPLS labels. The other example is of duplex on a
> Ethernet port. You can probably set the default to full duplex, and few
> folks will complain. If the protocol defines it and you can cite a section
> in the standard where the range/default is defined then you are covered.
>
> Where there is no clear standards defining ranges, defaults, min/max, I
> agree with you, it is best to keep it out of a standard model.
>
> Cheers.
>
> On May 27, 2015, at 10:53 PM, Anees Shaikh <[e-mail]> wrote:
>
> That said, where it's easy to add validation, we have tried to include it,
> though ranges, defaults, min/max values, etc. get harder when trying to
> build something vendor-neutral.
>
>
> Mahesh Jethanandani
> [e-mail]
>
>
>
>