On 09/02/2016 15:42, Kent Watsen wrote:
Can you please suggest an approach of how to return a single tree that
contains the data from two separate datastores (where the leaf paths may
overlap)? I think that the approach would need to work both for get
requests and also notification data.
I know that this is a difference between the solutions, but I don’t see it
listed as a requirement. There is a requirement to return the diff, but that’s
all. Is there actually a need to return both sets of data? - or is this just
a desire for the diff to be able to return both (as oppose to the transform)?
Just wondering if this is a must-have or a nice-to-have.
My desire is to find a overall solution that from a client usability
perspective is on a par with the OpenConfig models.
At the moment, my understanding is that the OpenConfig operators don't
have a need (or desire) to use datastores. I'm probably wrong, but
personally I think that for any datastore solution to really be viable
the use of those datastores needs to be made as seamless as possible for
the clients.
However, I still think that an encoding/protocol based solution has the
opportunity to provide a solution that from a client perspective feels
similar to the OpenConfig models, but has the additional benefit that it
doesn't force all the YANG models to have the intended/applied config
split, and arguably has the benefit that the source YANG models are
easier to read than the grouping approach used in the OpenConfig models.
Rob
Kent
_______________________________________________
netmod mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/netmod
