[nfsv4] Re: Draft adoption for WG - draft-dnoveck-nfsv4-security-10

[Christoph Hellwig <[e-mail]>]

On Wed, Jul 24, 2024 at 02:19:37PM -0400, David Noveck wrote:
> Right now a have to support it.   Right no way to do that is to simplify
> the task by dealing with complicated elements common to all minor
> versions in common NFSv4-wide documents.

Care to quote the IETF rules that require us to develop old and
mostly abandoned specs?

> > I think a big part of the problem is that it does not feel like we
> > a consensus on how to move forward.  Just to list what immediately comes
> > to mind:
> >
> >  - what status does 4.0 have for the working group?
> 
> 
> It's a protcol we own.  You might not like that but I don't see we have any
> choice in the matter.   There is no to say "Sorry Never Mind" for a
> Proposed Standard.   You may not like the decision for RFC5661 not to
> obsolete RFC3530 but it is toolate to change it now.

There are plenty of protocols in IETF that have not been obsoleted but
don't see ongoing development.

>  - how do we deal with pure errata for 4.1?  Are we doing to do a bis
> >    document for this big spec that does just errata?
> 
> 
> I assume so.  What else would we  do?

Well, the current bis document seems to be doing more than just
errata.  I just want to make sure we are all on the same page.

>  - how do we deal with behavior changes that are REQUIRED for actually
> >    making parts of 4.1 work.
> 
> 
> Do you know of any of these or os just speculation?

That's how 8881 happened.  If we don't want to go forward with more
of that that would suit me very well, but again I want WG consensus.

> Enhancements/extensions  are done in the nex endible minor ersion according
> to RFC8178.

That's not what for example your ACL document does, which targets
4.1 as far as I can tell.

> > Do they go into
> >    separate documents that apply to multiple versions?
> 
> 
> Probably not.

But that is what the internationalization and security documents do.


> > We'll need to reach and document working group consensus on these
> > high level decisions first before we go into new documents that
> > change the core protocol.
> >
> 
> What you seem to be talking about is a new document obsoleting and
> replacing RFC8178.

I don't think we are exactly following the document, which is
part of the issue.