Re: [storm] iSCSI version descriptors

Hi David (Harrington),

I agree with your summary, and I wanted to get your feedback on the
following approach for the iSCSI Consolidated spec.  

Unless I am seriously forgetting something, the nature of protocol changes
in the iSCSI Consolidated spec so far fall roughly into three categories:

1) Removing an optional feature
2) Removing non-mandatory negotiation options for a mandatory feature
3) Adding an optional feature

WG still has *not* come up with a need for the following:
4) Adding/Changing the mandatory behavior of protocol 
5) Removing a previously mandatory feature
6) Removing previously mandatory negotiation options for a mandatory feature

I do not believe changes of categories #1, #2 and #3 require a version
number bump-up - consistent with your discussion below.   So at the moment,
IMO, there's only need for 3 version descriptors:

0: no version claimed
1: RFC 3720+5048 through RFC-consolidated
2: RFC iSCSI-SAM

However, *should* the WG decide to make any changes in categories #4, #5 or
#6 down the road, I submit that (the eventual) RFC-consolidated will require
its own version number because that will be a semantic change in a mandatory
portion of the protocol.  So at that point, IMO, we should switch to 4
version descriptors:

0: no version claimed
1: RFC 3720+5048 
2: RFC-consolidated
3: RFC iSCSI-SAM

Comments/corrections are welcome.

Thanks.

Mallikarjun

  -----Original Message-----
  From: storm-bounces@ietf.org [mailto:storm-bounces@ietf.org] On Behalf
  Of David Harrington
  Sent: Wednesday, December 08, 2010 7:01 AM
  To: david.black@emc.com; paul_koning@Dell.com
  Cc: storm@ietf.org
  Subject: Re: [storm] iSCSI version descriptors

  Hi,

  Jumping in late ...

  I think compliance requirements are key here.
  If you can be "compliant to version 1" if you implement mandatory feature
  X AND if you do NOT implement mandatory feature X, then you don't have
  a very good compliance requirement.
  The purpose of a version # is to help identify the set of features that
can be
  expected to be supported by OTHER implementations, to ensure
  interoperability.
  (obviously, if the feature was originally, and continues to be, protocol-
  negotiable that's not a problem within a version#)

  I think it could be a poor engineering choice to have version 1 refer to
  multiple different possible sets of features. I disagree that "Yes, using
the
  same version should be a goal", if the set of features one implementation
  can expect of another implementation changes; interoperability should be
  the goal. You can ignore my engineering advice (I assume the WG knows the
  technology better than me). But I would appreciate being educated why this
  is OK.

  Interoperability is considered during IESG Evaluation
  http://www.ietf.org/iesg/statement/discuss-criteria.html. If I question
the
  interoperability of using the same version #, others in the IESG also
might
  question this. So you should ***document in each
  draft*** why using the same version # to identify the changes proposed in
  that draft is OK technically from an interoperability standpoint.

  (p.s. "we've already implemented it that way from the draft" is NOT an
  argument that holds much weight with the IESG).

  David Harrington
  Director, IETF Transport Area
  ietfdbh@comcast.net (preferred for ietf)
  dbharrington@huaweisymantec.com
  +1 603 828 1401 (cell)

  > -----Original Message-----
  > From: storm-bounces@ietf.org [mailto:storm-bounces@ietf.org] On
  Behalf
  > Of david.black@emc.com
  > Sent: Monday, October 25, 2010 9:56 AM
  > To: paul_koning@Dell.com
  > Cc: storm@ietf.org
  > Subject: Re: [storm] iSCSI version descriptors
  >
  > > For now, I believe we should say 3270/5048 and not mention
  > "new consolidated draft".  It doesn't
  > > seem appropriate to reference a draft.  If we end up being
  > able to reuse the code for the
  > > consolidated spec, as intended, then at the time that spec
  > comes out, the description of the code
  > > points would be updated to say that it applies to that RFC as
  well.
  >
  > Yes, the specifications of the version code values need to be in terms
  > of RFC numbers, not references to Internet-Drafts.
  >
  > Thanks,
  > --David
  >
  >
  > > -----Original Message-----
  > > From: Paul Koning [mailto:paul_koning@Dell.com]
  > > Sent: Friday, October 22, 2010 11:14 AM
  > > To: Black, David
  > > Cc: storm@ietf.org
  > > Subject: Re: [storm] iSCSI version descriptors
  > >
  > >
  > > On Oct 22, 2010, at 1:14 AM, <david.black@emc.com>
  > <david.black@emc.com> wrote:
  > >
  > > > One more time on this issue.  This is for discussion -
  > it's not an announcement of a decision.
  > > This is the only real issue that needs to be resolved to
  > complete the new (SAM) features draft for
  > > iSCSI.
  > > >
  > > > Reminder: we need to define a set of small positive
  > integer values to describe the iSCSI version
  > > starting with 0 = "no version claimed".  After some private
  > discussions, it appears that we need two
  > > additional version values beyond 0.
  > > >
  > > > The first observation is that the baseline should be at
  > least RFC 3720 (original iSCSI) + RFC 5048
  > > (Corrections and Clarifications).  That would be version value 1.
  > > >
  > > > The next observation is that taking features out of the
  > consolidated iSCSI draft may allow a
  > > visible behavior change.  RFC 3720 has this to say about
  > text keys for negotiation:
  > > >
  > > >   All keys in this document, except for the X extension
  > formats, MUST
  > > >   be supported by iSCSI initiators and targets when used
  > as specified
  > > >   here.  If used as specified, these keys MUST NOT be
  > answered with
  > > >   NotUnderstood.
  > > >
  > > > When we take out a feature in the new iSCSI consolidated
  > draft, the easiest thing to do is allow a
  > > NotUnderstood response to the keys that negotiate that
  > feature.  This should not pose a problem for
  > > unimplemented features, but it would be a behavior change.
  > The completely backwards-compatible
  > > alternative is have the consolidated iSCSI draft list the
  > keys used for removed features and
  > > prohibit  a NotUnderstood response to those keys (Reject
  > would be an acceptable alternative
  > > response).
  > >
  > > That sounds right.  Does it cause problems for that to be
  > the answer?
  > >
  > > > If we're careful about this, the same version value can
  > apply to 3720/5048 and the consolidated
  > > iSCSI draft.  I'd suggest that we be careful, and the
  > details of how can be worked out as we
  > > finalize the consolidated draft - I think we should have at
  > least one more round of looking at
  > > features to remove.
  > >
  > > Yes, using the same version should be a goal.
  > >
  > > > After that, we'll need a version value for the new (SAM)
  > features draft additions.  The result
  > > would be 3 version values:
  > > > 	0 = no version claimed
  > > > 	1 = 3720/5048 or new consolidated draft
  > > > 	2 = (3720/5048 or new consolidated draft) + SAM 4/5
  > features from the SAM draft.
  > >
  > > I assume these numbers will be handled by IANA, so they'll
  > need an IANA Considerations section
  > > spelling out how.
  > >
  > > For now, I believe we should say 3270/5048 and not mention
  > "new consolidated draft".  It doesn't
  > > seem appropriate to reference a draft.  If we end up being
  > able to reuse the code for the
  > > consolidated spec, as intended, then at the time that spec
  > comes out, the description of the code
  > > points would be updated to say that it applies to that RFC as
  well.
  > >
  > > For the SAM draft, can we reference a specific draft?  I
  > think it's quite absurd that drafts even
  > > have standard status in the first place, but given that
  > they do, we need to be precise about which
  > > document we're talking about.  If we say "draft 2010/10/34"
  > now, and then a followup draft comes out
  > > for which the same code point is appropriate, we'd update
  > the description of the code point to list
  > > that one as well.  But we can't leave it unstated.
  > >
  > > 	paul
  > >
  >
  > _______________________________________________
  > storm mailing list
  > storm@ietf.org
  > https://www.ietf.org/mailman/listinfo/storm
  >

  _______________________________________________
  storm mailing list
  storm@ietf.org
  https://www.ietf.org/mailman/listinfo/storm