Andy Bierman <a...@yumaworks.com> writes: > On Thu, Apr 28, 2016 at 12:15 PM, Juergen Schoenwaelder < > j.schoenwael...@jacobs-university.de> wrote: > >> On Mon, Mar 07, 2016 at 11:26:36AM +0100, Juergen Schoenwaelder wrote: >> > On Tue, Feb 16, 2016 at 12:42:27PM +0100, Juergen Schoenwaelder wrote: >> > > So perhaps the proposal is to add >> > > >> > > After applying all deviations announced by a server, in any order, >> > > the resulting data model MUST still be valid. >> > > >> > > just before the beginning of section 7.20.3.1? >> > >> > I like to see whether we have consensus to add this clarifying >> > sentence. I believe it documents something that we always assumed to >> > be true but we did not write it down explicitly. If anyone disagrees >> > with adding this clarification, please speak up now. >> > >> >> I have not heard anyone disagreeing with adding this clarification. >> >> Martin did not put this additional sentence into -12 and since we seem >> to have consensus on adding this clarification I like to ask him to >> add this sentence if another revision is made before the IESG >> telechat. I am CCing Benoit in case he wants to add this as a comment >> to his IESG evaluation record. >> >> > Was it unclear before that the result of deviations were allowed to be > invalid?
Right, I think it goes pretty much without saying, so this sentence is IMO unnecessary. The real problem is that different ways of applying the deviations may lead to different data models, all valid. I think this should be avoided although I don't know how to do it. In hindsight, it would probably be better to allow the server to specify just a single deviation module with no other content. Then it would be easier to state some rules. > I do not object to this text, but I hope those 3 words "in any order" are > enough > of a clue to developers that on the server "only one wins" but the client > does not know which one. I will add a YANG guideline to avoid duplicate > deviation targets. (issue #33) I don't think it is clear that only one wins. Three deviations adding default values to the same leaf-list target could be possibly applied together. Lada > > > > /js >> >> > > Andy > > > >> -- >> Juergen Schoenwaelder Jacobs University Bremen gGmbH >> Phone: +49 421 200 3587 Campus Ring 1 | 28759 Bremen | Germany >> Fax: +49 421 200 3103 <http://www.jacobs-university.de/> >> >> _______________________________________________ >> netmod mailing list >> netmod@ietf.org >> https://www.ietf.org/mailman/listinfo/netmod >> > _______________________________________________ > netmod mailing list > netmod@ietf.org > https://www.ietf.org/mailman/listinfo/netmod -- Ladislav Lhotka, CZ.NIC Labs PGP Key ID: E74E8C0C _______________________________________________ netmod mailing list netmod@ietf.org https://www.ietf.org/mailman/listinfo/netmod
