Re: [nfsv4] draft haynes-nfsv4-versioning: Issues relating to opcode assignments etc.

[David Noveck <[e-mail]>]

> There are a number of allocation policies that you can use.

Perhaps so, but we still have the issue of whether there is a reason to use
them, i.e. whether they add anything to the process., other than an extra
interaction with IANA.

> You can have a first-come-first-served-with-Expert-review process,

I don't think you can combine FCFS with expert-review in that way, and
have things work..

For example, RFC5661, specified, for the named attribute registry, both
FCFS and specification-required.  That got thorough IESG and IANA review
and RFC%661 was approved as a Proposed Standard.  So far so good, but ...

rfc3530bis dutifully copied the language of RFC5661 (since we wanted to be
compatible with RFC5661 on this point).  When the bis document went through
the review process, we ran into a brief period of IANA-NOT-OK and IANA
explained to us that we had to choose one policy or the other.

> which is what I would recommend for any namespaces large enough that
you're not going to
> worry about running out.

The problem is that  the issues of holes in the allocations, that Benny was
concerned about becomes magnified.  Even  if you aren't worried about
an occasional unassigned attribute or operation code, this policy could
create large spaces in the allocations.  For example, if someone assigned
attribute number 10,000, some GETATTR's might be thousands of bytes long.
Given current implementations, which we don't want to change radically,
running out of opcodes/attributes is not the only issue.

On Tue, Oct 28, 2014 at 12:01 PM, Nico Williams <[e-mail]>
wrote:

> On Tue, Oct 28, 2014 at 11:23:30AM -0400, David Noveck wrote:
> > I think we need to distinguish a couple of related issues:
> >
> >    - How (or by whom) the assignment is made.
> >    - When the assignment is made
> >
> > I think the issue about which there is likely to be disagreement or
> > controversy will be the question of when (i.e. how early) the assignment
> is
> > to be made. My understanding of the connection between these two issues
> is
> > as follows:
> >
> >    - If the assignment is to be made when the RFC is published, then we
> >    could implement IANA-based assignment by creating one or more
> registries
> >    with a "specification required" assignment policy.  However, this
> could be
> >    complicated to do, although it is relatively simple if you just do it
> for
> >    operations.  The big issue is that doing this doesn't buy you
> anything (see
> >    below for a discussion of this).
> >    - If the assignment is to made earlier than that (e.g. when the
> document
> >    becomes a working-group document as suggested in the current draft),
> it
> >    isn't clear how to specify an assignment policy that IANA would
> accept.
>
> There are a number of allocation policies that you can use.  Having a
> Protocol Action from the IESG is not the only way.  You can have a
> first-come-first-served-with-Expert-review process, which is what I
> would recommend for any namespaces large enough that you're not going to
> worry about running out.
>
> Nico
> --
>