Re: [yang-doctors] RFC 9127-bis question - impact of reference clauses in an import statement

[Martin Björklund <[e-mail]>]

Hi,

Jeffrey Haas <[e-mail]> wrote:
> Dear YANG doctors,
> 
> A wonderful end of year holiday gift would be to ship the fix to RFC 9127.
> 
> Please answer the question below.
> 
> -- Jeff
> 
> On Thu, Dec 09, 2021 at 05:12:07PM -0500, Jeffrey Haas wrote:
> > [Reference message to issue in thread:
> > https://mailarchive.ietf.org/arch/msg/rtg-bfd/Ajo-ku8EJxiEnZjdkRfdzGIg59A/ ]
> > 
> > YANG doctors,
> > 
> > The BFD Working Group is trying to move forward on the RFC 9127-bis work
> > that was previously discussed with you.  I believe there is consensus in how
> > we will move forward on the modification to the common configuration
> > grouping and its associated feature statement.  This strategy was previously
> > communicated to you all.
> > 
> > This is a question about what we put in the -bis document.  Previously, the
> > thinking was we did a full re-issue.  Tom Petch, in the thread cited above,
> > notes that we minimally can't do part of this because there are modules
> > already delegated to IANA.
> > 
> > The discussion about the what the internet-draft text should include from
> > RFC 9127 or not lead to a question I had: Would it not be sufficient to
> > simply issue an update to the ietf-bfd-types module where all of the impact
> > is isolated?
> > 
> > Tom's argument is that the form of the import, which contains references to
> > RFC 9127, suggest we really want a re-issue so we point to the proper new RFC.
> > 
> > As an example, from module ietf-bfd:
> >  :  import ietf-bfd-types {
> >  :    prefix bfd-types;
> >  :    reference
> >  :      "RFC 9127: YANG Data Model for Bidirectional Forwarding
> >  :       Detection (BFD)";
> >  :  }
> > 
> > As best I am able to tell from RFC 7950, § 7.21.4, reference is partially
> > meant to be informational.  If it was normative, an update to any module
> > imported by another with a reference would require an update.

Correct.

> > Note that this isn't an import by revision.
> > 
> > My question is thus what are the rules to issue module updates in such
> > circumstances?

In the general case this happens as soon as you publish an update to
an published module that is already used.  This is fine.

What makes this case special is that you plan to (re)publish modules
with references to the document that you are replacing.

I'm not sure that this is a YANG doctors question... it seems more of
a procedural issue.

> > 
> > Tom and the BFD YANG authors are copied to make sure I haven't
> > misrepresented the situation.
> > 
> > -- Jeff

I noted in the diff that Mahesh sent that you have:

  revision 2021-12-07 {	
    description	
      "9127-bis.";	
    reference	
      "RFC XXXX: YANG Data Model for Bidirectional Forwarding	
                 Detection (BFD).";	
  }

The description needs to be updated.

Also, the description statement in ietf-bfd-types should be updated to
use the new TLP (with "Revised BSD" instead of "Simplified BSD").

... but then the rest of the modules will have the old language.  I
don't know if this is acceptable or not.


Would it be possible to avoid the problem by not doing a full
9127bis, but just publish an RFC with a new version of ietf-bfd-types?



/martin