Re: [core] Robert Wilton's Discuss on draft-ietf-core-sid-16: (with DISCUSS and COMMENT)

[Michael Richardson <[e-mail]>]

Rob Wilton (rwilton) <[e-mail]> wrote:
    > leaf foo {
    > type string {
    > len "1..20";
    > }
    > }

    > Or the definition of foo could change in a non-backwards-compatible way.  E.g.,

    > leaf foo {
    > type int32;
    > }

    > RFC 7950 basically states that this is not allowed, but these sorts of
    > changes happens anyway (e.g., vendors fixing bugs, or bugs in standard
    > YANG models), and the YANG versioning work will allow these sorts of
    > changes with appropriate mitigations.
    > https://datatracker.ietf.org/doc/html/draft-ietf-netmod-yang-module-versioning-03

    > But in all these cases, no new SID should be allocated, or otherwise
    > you will have two SIDs assigned to the same name/path.

Encoding towards CBOR, the CBOR wouldn't care that much about the change in
type, as the CBOR is self-describing in that way.
So maybe it's okay if a new SID is not allocated.

Where it would be a problem is if the change was from a 1-based count to a
0-based count, both encoded as integers.  In that case, I really hope it
would go from "foo" to "foo-0" or something like that.

    >> > The draft states that if the
    >> > definition is changed in a non-backwards-compatible (NBC) way then a
    >> > new SID
    >> > SHOULD be allocated.
    >>
    >> True, I assume that this leads to a new YANG leaf name.

    > Only if the author gives the definition a new name, and from a SID
    > allocation POV, this is just a new item name in the schema, and hence
    > needs a new SID.  I.e., the allocation of a new SID is related to a new
    > item name, not because of changes in a definition.

I agree that we have a problem enforcing at the pyang sid generation place
that non-backwards compatible changes need new SIDs.  That sounds like a
tooling improvement, rather than a bug in the spec.

    >> > The policy in 7.6 for the "IETF
    >> > YANG SID Registry" requires an RFC.  What is the mechanism if an
    >> > individual or
    >> > open source project wanted to get SIDs assigned for some of their YANG
    >> > modules?
    >>
    >> > I.e., should we be defining a separate mega-range, managed by IANA,
    >> > with just
    >> > Expert Review or Specification Required so that these modules could use
    >> > SIDs
    >> > allocated?  Or do you envisage a separate entity taking up the
    >> > responsibility
    >> > for coordinating this?
    >>
    >> My impression was that there would be a mega-range operated outside of
    >> IANA
    >> for this (open source, non-RFC YANG module) kind of thing, but I think that
    >> the energy for doing that may have waned.

    > Okay.  If there is a lack of energy there, do you think that this is
    > something that we should be asking IANA to manage (i.e., have this
    > draft separate a separate mega-range for these other modules)?

So, we should have a second mega-range allocation, with a FCFS
considerations.  I would be in favour of that.
(It could also happen in a new document or via some other IESG action.)

--
Michael Richardson <[e-mail]>   . o O ( IPv6 IøT consulting )
           Sandelman Software Works Inc, Ottawa and Worldwide