On 08/12/2021 13:38, Ladislav Lhotka wrote:
tom petch <[email protected]> writes:
The BFD WG are revising RFC9127 to add a new feature if-feature
"client-base-cfg-parms"; and make uses base-cfg-parms { conditional
thereon in module ietf-bfd-types. Reading and re-reading RFC7950,
especially about mandatory and top-level, I am not convinced that
this is legal.
Sorry, I don't get the problem - nothing in the "base-cfg-parms" grouping is
mandatory. Why do you think this might be illegal?
RFC7950, Section 11 (Updating a Module). A change like this does not
fall under any of the allowed revisions of definitions in a published
module, therefore:
Otherwise, if the semantics of any previous definition are changed
(i.e., if a non-editorial change is made to any definition other than
those specifically allowed above), then this MUST be achieved by a
new definition with a new identifier.
The semantics of the original "uses" (the one Tom mentioned) change if
an "if-feature" is added to it. You start with a reusable set of data
definition nodes and end up with a reusable set of data definition nodes
that are constrained by some condition.
Jernej
Lada
The module bfd-types is imported by a number of other modules such
as OSPF, RIP, PIM so it is also a question if e.g. a leaf can be
made mandatory by its usage in another module. I raised this on the
BFD list and the WG Chair tells me that this is a violation of the
intent of the RFC, 7950, but that it has been reviewed by YANG
doctors and is probably the best fix.
If YANG Doctors collectively say that this violation is ok, then I think that
such a statement needs to appear on the Netmod WG list.
I think that there are a lot of other editorial changes needed to 9127-bis to
make it legal but they can come later. The I-D is in WG Last Call ending
20Dec2021
Tom Petch
_______________________________________________
netmod mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/netmod
_______________________________________________
netmod mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/netmod
