Hi Jeff, Yes, to me it makes sense to do the change suggested by Martin (add "default tx-rx-intervals;" to the choice statement). BFD YANG co-authors, please respond asap if you disagree.
Regards, Reshad. On 2019-08-21, 4:11 PM, "Jeffrey Haas" <[email protected]> wrote: Reshad, If procedures permit it (I'm unclear on the detail), does it make sense to pull the BFD yang module for a fix from the editor queue? -- Jeff On Mon, Aug 19, 2019 at 07:31:27PM +0000, Reshad Rahman (rrahman) wrote: > I was looking at an old copy of the doc which didn't have default. So yes, mandatory doesn't make sense with the default statements. > > Your assumption below wrt the intention is correct. I don't know how feasible it is to add this while it's in the editor q. > > Regards, > Reshad. > > On 2019-08-19, 3:18 PM, "Martin Bjorklund" <[email protected]> wrote: > > "Reshad Rahman (rrahman)" <[email protected]> wrote: > > Thanks Martin and Mahesh. > > > > I believe we should add a mandatory statement to the choic (speaking > > as BFD YANG co-author,) > > But then it is not clear why all leafs in the cases have default > statements. > > Since the 'single-interval' case is optional with a if-feature (which > BTW is weird since it is trivial to implement), and the only other > case has default values on both its leafs, I would have assumed that > the intention was that if nothing is configured, the server should use > 1000000 microseconds for the intervals. If this is the intention, > perhaps a statement: "default tx-rx-intervals;" can be added to the > module, even though the doc is in the RFC ed q. > > > /martin > > > > > > > Just created https://github.com/bfd-wg > > > > Regards, > > Reshad. > > > > > > On 2019-08-19, 2:45 PM, "Mahesh Jethanandani" <[email protected]> wrote: > > > > [Adding the authors of BFD YANG module] > > > > Martin brings up a good point. But since the document that contains ietf-bfd-types is sitting in RFC Ed Queue, this will have to go into a bis document. > > > > Chairs, could you create a bfd-wg in GitHub for us to track this as an issue to be fixed as part of a bis document? > > > > > On Aug 19, 2019, at 4:29 AM, Martin Björklund via Datatracker <[email protected]> wrote: > > > > > > Reviewer: Martin Björklund > > > Review result: Ready with Nits > > > > > > I have reviewed this document from a YANG model perspective only. > > > > > > My only comment is actually for a grouping defined in ietf-bfd-type, but used > > > in this module. There is a choice "interval-config-type": > > > > > > +--rw unsolicited {bfd-unsol:unsolicited-params-global}? > > > +--rw enable? boolean > > > +--rw local-multiplier? multiplier > > > +--rw (interval-config-type)? > > > +--:(tx-rx-intervals) > > > | +--rw desired-min-tx-interval? uint32 > > > | +--rw required-min-rx-interval? uint32 > > > +--:(single-interval) {single-minimum-interval}? > > > +--rw min-interval? uint32 > > > > > > This choice is not mandatory and doesn't have a default case, so the question > > > is what happens if no nodes from the choice has been configured? I would > > > expect the choice to have a default case (but this then would apply to > > > ietf-bfd-types, not this document.) > > > > > > > > > > Mahesh Jethanandani > > [email protected] > > > > > > > > > > > >
