Med, On Mon, Aug 30, 2021 at 12:19:23PM +0000, mohamed.boucad...@orange.com wrote: > Hi Greg, > > Thank you for checking the OAM part and for sharing this comment. > > As you can read in both sections 4 and 5, this model is ** not a device > configuration model **. The focus is on aspects that can be triggered by > service requests and managed by the network controller. This network view of > the service will be then enriched (with other sources such as local > templates/profiles/defaults) to derive the exhaustive configuration that will > be enforced in involved devices to deliver the requested service. > > With that rationale in mind, you can understand why we don’t import device > models but point to the authoritative RFCs for aspects that we think make > sense to be tweaked at the network-level.
Greg's suggestion is not to import a device specific model. His suggestion is to leverage groupings defined in the BFD yang model that have the configuration semantics typical of most BFD implementations. Put a different way, if you're defining BFD parameters, why would you have different names or structures for BFD configuration elements in your module than everyone else? -- Jeff
