Re: [AAA-DOCTORS] [OPS-DIR] FW: Internal WG Review: STORage Maintenance (storm)

["David Harrington" <[e-mail]>]

Hi,
inline. 

> -----Original Message-----
> From: [e-mail] [mailto:[e-mail]] 
> Sent: Thursday, April 30, 2009 7:02 PM
> To: [e-mail]; [e-mail]; [e-mail]
> Cc: [e-mail]; [e-mail]; [e-mail]; 
> [e-mail]; [e-mail]
> Subject: RE: [OPS-DIR] FW: Internal WG Review: STORage 
> Maintenance (storm)
> 
> Let me start by acknowledging that David Harrington is right
> about the absence of operations and management in the charter.
> At a minimum, the new iSCSI features strongly suggest an
> updated version of the iSCSI MIB (RFC 4544), and I need to
> revise the draft charter to include that.  If David Harrington
> is willing to serve as a technical advisor to facilitate this
> update, that would be helpful.

Part of my concern is whether SNMP MIBs are the correct answer for
storage management. 
I do not know the answer to the question. Much of the charter
discusses bringing the 
IETF storage standards into compliance with implementation/deployment
reality.

I think the WG should also consider how to bring the storage MIBs into
compliance with
implementation/deployment reality. At a minimum, if new features are
added to the IETF storage
stanards, then the MIBs should also be updated. (I'll check with my
bosses to see if
they will support my acting as MIB/OPS technical advisor.)

But if the storage MIBs are not being used, and other management
standards are used for managing storage,
then maybe the IETF needs to acknowldge that reality and have our
standards accommodate that.

> 
> OTOH, I disagree with the suggestion that the WG "consider
> and document" ... "what changes might need to be made to
> utilize the existing non-IETF standards to manage the updated
> IETF storage standards".  IMHO, it is usually better to
> communicate the functional changes that the IETF is making
> in its protocols to other standards organizations that
> are responsible for related (e.g., management) standards.
> That enables those organizations to make any changes that
> they deem appropriate based on their expertise - in contrast,
> issuing explicit instructions to another standards organization
> has the potential to go over badly (IMHO).  The draft charter
> already contains references to working relationships with T10
> and T11 - it looks like it needs corresponding text for SNIA
> around the iSCSI HBA API and SMI-S support for iSCSI.

I think you misunderstood my intentions. If non-IETF standards for
management 
(SNIA SMI-S and T10/T11 HBA) are widely used in the real world, and
IETF management
standards are not widely used to manage iSCSI, FCIP, etc., then we
should 
consider how **our** standards - iSCSI, FCIP, storage MIBs, etc.,
should be
modified to work better with those widely-deployed non-IETF standards
for 
storage management. Or whether a combination of IETF management
standards (syslog,
netconf, and SNMP) would be more suitable than the SNMP-only approach.

If SCSI is usually managed using HBA and not SNMP, then we should
consider whether
**i**SCSI needs to be changed to allow consistent (both i* and non-i*)
SCSI management.
Maybe we need changes to iSCSI to make it compatible with SNMP and HBA
and SMI-S.

> 
> Not announcing the BoF on the IETF list was my oversight;
> that clearly should have been done prior to San Francisco.
> I'll work with Lars on ensuring that there is sufficiently
> broad IETF exposure and review of the proposed storm charter.

Thanks.

> 
> For the security concerns, please see RFC 3723 (Securing
> Block Storage Protocols over IP).  I expect that RFC to
> remain applicable to all of the revised protocols, with the
> overall approach being that security issues in technology that
> is standardized by another standards body are the responsibility
> of that standards body; would a statement to that effect (RFC
> 3723 expected to remain applicable, security issues in standards
> from other organizations remain the responsibility of those
> organizations) in the charter be helpful?  

Having such a statement would have made it clearer to people like me
who are not in the other organizations.

> 
> That said, I am surprised by what appears to be an inference
> that lack of mention of security in the charter implies that
> security may not be addressed at all.  All RFCs have to address
> security considerations, making security something that every
> IETF WG has to address, independent of whether security is
> explicitly mentioned in the WG charter.  If every WG charter
> now has to have a "Security Considerations" section, there are
> number of WG charters on the IETF web site that need updates
> ... surely the IESG has better things to do ;-).

I did not overlook that you would need to deal with Security
Considerations.

I did overlook RFC3723. There are lots of documents related to iSCSI 
and iFC and FCIP and iSNS and ... I simply haven't had time to study
all of them.

The problem I have is that I could not look at the documents from
INCITS 
to determine the status of security for SCSI/iSCSI and FC/FCIP, etc. 
I suspect the INCITS documents do not directly discuss iSCSI security.
My 
impression is T10/T11 focus on non-Internet connectivity, where
security
concerns/threats might be different than for Internet connectivity.

INCITS did make public their intention to update SAM-4 to address
security.
Apparently they think there is something that needs addressing.
But I don't know what they know, because they don't say what needs 
addressing and you don't say anything about upgrading security as you
upgrade the i* variations of the underlying protocols.
(Maybe the SAM-4 pretty picture of the architecture needs to be
updated
to show the security boxes. But the SAM-5 proposal simply doesn't say
what 
needs to be done, so i cnanto see if it is something simple or
something difficult
that could impact making storage accessible securely over the
Internet.)

> 
> Regarding SAM-5, the draft storm WG charter references SAM-4
> and not SAM-5 for a reason; SAM-5 is an out-of-scope moving
> target.  

Understood. I was not suggesting we should try for SAM-5
compatibility.

> FWIW, most SCSI (T10) security is not in the SAM
> standards, but rather in the SPC-4 standard (under development).
> For Fibre Channel (T11) security, see the FC-SP standard and
> the FC-SP-2 standard (under development).
> 
> That brings me to the T10 and T11 standards access concerns,
> which are real and something I need to work on.  INCITS, the
> parent organization of T10 and T11, imposed new document
> access restrictions within the past year.  I need to work
> with T10 and T11 on how to go about making the necessary
> standards documents available to IETF participants, starting
> with T10 (which meets next week).  I do not intend to ask
> the IESG for approval of the storm WG charter until the
> document access concerns are resolved.
> 
> Thanks,
> --David
>  
> 
> > -----Original Message-----
> > From: David Harrington [mailto:[e-mail]] 
> > Sent: Thursday, April 30, 2009 12:30 PM
> > To: 'Lars Eggert'; 'Ersue, Mehmet (NSN - DE/Munich)'
> > Cc: [e-mail]; [e-mail]; Black, David; 
> > [e-mail]; [e-mail]; 'IAB'
> > Subject: RE: [OPS-DIR] FW: Internal WG Review: STORage 
> > Maintenance (storm)
> > 
> > Hi,
> > 
> > I found that having the discussion for the STORM BOF on the
mailing
> > lists for (three) concluded WGs made it a bit obscure.
> > As far as I can tell by looking at those archives, there wasn't
much
> > discussion.
> > I will also note that the announcement of the craetion of the
storm
> > list seems to be have been only announced to the lists of the
three
> > concluded WGs. So if you weren't involved in the three earlier
WGs,
> > you might not have seen the announcement. 
> > 
> > I do think Mehmet has a point that many IETF people may not 
> have been
> > aware the discussion was happening.
> > 
> > As an OPSDIR member and MIB Doctor, I have already 
> expressed concerns
> > that operations and management is not being addressed in 
> the charter.
> > The chair helpfully responded with a short list of management
> > standards (SNIA SMI-S, SNIA SCSI HBA, INCITS FC HBA) that might
help
> > answer my questions. Unfortunately, I cannot access some of the
> > documents referenced because they are not freely available. 
> If I read
> > through a number of large standards I might be able to determine
for
> > myself what changes might need to be made to utilize the existing
> > non-IETF standards to manage the updated IETF storage standards. I
> > think I would rather see the WG consdier this and document 
> it, rather
> > than me (and any other user of the IETF technologies) having to
> > research this personally to know the answers. I would nto say my
> > concerns have been addressed.
> > 
> > As a SECDIR member, I have concerns that security may not be
gettng
> > addressed either, but it is hard for me to say that for sure since
I
> > cannot freely access the T10 and T11 documents that are the basis
of
> > the updates. I do not know whether the updated IETF storage 
> standards
> > will utilize IETF standards for security or non-IETF stndards for
> > security, or whether the WG is simply not going to address
security
> > issues since it is not mentioned in the charter. I note that the
> > charter calls for updating to SAM-4 level, and that T10 is
starting
> > new work on SAM-5, explicitly to address security. So does that
mean
> > security will not be addressed in the STORM WG?
> > 
> > I have not seen these discussions elsewhere. And Mehmet's concerns
> > seem to be relevant to that point.
> > 
> > Before the WG is created, I think the charter should be 
> explicit about
> > the what the WG will do to address how these Internet 
> protocols should
> > be operated and managed and secured.
> >  
> > David Harrington
> > [e-mail]
> > [e-mail]
> > [e-mail]
> > 
> > > -----Original Message-----
> > > From: [e-mail] 
> > > [mailto:[e-mail]] On Behalf Of Lars Eggert
> > > Sent: Tuesday, April 28, 2009 12:44 PM
> > > To: Ersue, Mehmet (NSN - DE/Munich)
> > > Cc: [e-mail]; [e-mail]; 
> > > [e-mail]; [e-mail]
> > > Subject: Re: [OPS-DIR] FW: Internal WG Review: STORage 
> > > Maintenance (storm)
> > > 
> > > Hi,
> > > 
> > > On 2009-4-28, at 12:29, Ersue, Mehmet (NSN - DE/Munich) wrote:
> > > > unfortunately I missed the Storm BoF in SF. I also was not
aware
> > of
> > > > the discussion on the maillists you mention.
> > > >
> > > > I made the experience that announcing a BoF maillist to IETF
and
> > > > forcing some discussion on this maillist with IETF member  
> > > > participation
> > > > is a good way to show IESG that there is measurable 
> IETF community
> > > > interest.
> > > 
> > > as David said below, there was discussion on the 
> > > still-existing RDDP,  
> > > IPS and IMSS lists, and the BOF went - in my opinion - 
> very well. I
> > 
> > > believe community interest has been demonstrated, which 
> is why we're
> > 
> > > going forward with a charter proposal.
> > > 
> > > > The huge amount of people in the session, which are sometimes
> > > > occasionally in the room and vote, usually disappear when 
> > > it comes to
> > > > maillist discussion. Also for our regular WG work the
decisions
> > have
> > > > to be confirmed on the maillist.
> > > >
> > > > As I said I believe this is an important and necessary
> > consolidation
> > > > work but should be discussed and accepted first in the IETF 
> > > community.
> > > 
> > > I don't quite know what to do with you comment. Are you 
> saying there
> > 
> > > hasn't been sufficient discussion to go forward with the 
> chartering?
> > 
> > > Or am I misunderstanding?
> > > 
> > > I'll also note that the community obviously still has time to 
> > > comment  
> > > - the charter is in internal I* review at this time, 
> after which it
> > 
> > > will go for public review.
> > > 
> > > Lars
> > > 
> > > >
> > > > Cheers,
> > > > Mehmet
> > > >
> > > >
> > > >> -----Original Message-----
> > > >> From: ext [e-mail] [mailto:[e-mail]]
> > > >> Sent: Tuesday, April 28, 2009 2:32 PM
> > > >> To: Ersue, Mehmet (NSN - DE/Munich)
> > > >> Cc: [e-mail]; [e-mail]
> > > >> Subject: RE: [OPS-DIR] FW: Internal WG Review: STORage
> > > >> Maintenance (storm)
> > > >>
> > > >> Mehmet,
> > > >>
> > > >> The storm list is completely new (it was only opened for
people
> > > >> to subscribe within the past few days).  The storm BOF in San
> > > >> Francisco was organized using the existing IPS, RDDP and IMSS
> > > >> mailing lists.  I suggest consulting the archives for those
> > > >> lists, particularly for the IPS list, where you should find a
> > > >> reasonable level of community interest.
> > > >>
> > > >> There is definite community interest in this work, and I have
> > > >> author commitments for drafts for all six work items listed
> > > >> in the charter - each of the "First version of" milestone
> > > >> dates in the proposed charter is based on discussion with an
> > > >> author or authors who believe that a first version of the
draft
> > > >> will be ready by that date.
> > > >>
> > > >> Thanks,
> > > >> --David
> > > >> ----------------------------------------------------
> > > >> David L. Black, Distinguished Engineer
> > > >> EMC Corporation, 176 South St., Hopkinton, MA  01748
> > > >> [phone]             FAX: [phone]
> > > >> [e-mail]        Mobile: [phone]
> > > >> ----------------------------------------------------
> > > >>
> > > >>> -----Original Message-----
> > > >>> From: Romascanu, Dan (Dan) [mailto:[e-mail]]
> > > >>> Sent: Tuesday, April 28, 2009 6:33 AM
> > > >>> To: Ersue, Mehmet (NSN - DE/Munich)
> > > >>> Cc: Black, David
> > > >>> Subject: RE: [OPS-DIR] FW: Internal WG Review: STORage
> > > >>> Maintenance (storm)
> > > >>>
> > > >>> Mehmet,
> > > >>>
> > > >>> David Black ran the BOF in San Francisco. He should be able
> > > >> to provide
> > > >>> you information about the preparation work, level of 
> support and
> > > >>> interest from the community and initial drafts.
> > > >>>
> > > >>> Thanks for looking into this and for asking the questions.
> > > >>>
> > > >>> Dan
> > > >>>
> > > >>>
> > > >>>> -----Original Message-----
> > > >>>> From: Ersue, Mehmet (NSN - DE/Munich)
> > > >> [mailto:[e-mail]]
> > > >>>> Sent: Tuesday, April 28, 2009 1:29 PM
> > > >>>> To: Romascanu, Dan (Dan)
> > > >>>> Subject: RE: [OPS-DIR] FW: Internal WG Review: STORage
> > > >>>> Maintenance (storm)
> > > >>>>
> > > >>>>
> > > >>>> Hi Dan,
> > > >>>>
> > > >>>> I believe this is an important and necessary consolidation
> > work.
> > > >>>>
> > > >>>> I might have missed the discussion on another maillist but
> > > >>>> the storm maillist is pretty much new and I would have a
> > > >>>> better feeling if there were some mail traffic showing the
> > > >>>> interest of the community and/or an initial draft with an
> > > >>>> issues list for the justification of a new WG.
> > > >>>>
> > > >>>> Cheers,
> > > >>>> Mehmet
> > > >>>>
> > > >>>>
> > > >>>>> -----Original Message-----
> > > >>>>> From: [e-mail]
> > > >>>>> [mailto:[e-mail]] On Behalf Of ext 
> Romascanu,
> > > >>>>> Dan (Dan)
> > > >>>>> Sent: Monday, April 27, 2009 7:57 PM
> > > >>>>> To: [e-mail]; [e-mail]
> > > >>>>> Subject: [OPS-DIR] FW: Internal WG Review: STORage
> > > >>>> Maintenance (storm)
> > > >>>>>
> > > >>>>>
> > > >>>>>
> > > >>>>>
> > > >>>>> -----Original Message-----
> > > >>>>> From: [e-mail] 
> [mailto:[e-mail]] On
> > > >>>>> Behalf Of
> > > >>>>> IESG Secretary
> > > >>>>> Sent: Monday, April 27, 2009 8:54 PM
> > > >>>>> To: [e-mail]; [e-mail]
> > > >>>>> Cc: [e-mail]
> > > >>>>> Subject: Internal WG Review: STORage Maintenance (storm)
> > > >>>>>
> > > >>>>> A new IETF working group is being considered in the 
> Transport
> > > >>>>> Area.  The
> > > >>>>> draft charter for this working group is provided below for
> > > >>>> your review
> > > >>>>> and comment.
> > > >>>>>
> > > >>>>> Review time is one week.
> > > >>>>>
> > > >>>>> The IETF Secretariat
> > > >>>>>
> > > >>>>> STORage Maintenance (storm)
> > > >>>>> ----------------------------------
> > > >>>>>
> > > >>>>> Last Modified: 2009-04-25
> > > >>>>>
> > > >>>>> Current Status: Proposed Working Group
> > > >>>>>
> > > >>>>> Chairs:
> > > >>>>> - David L. Black <[e-mail]>
> > > >>>>> - tbd
> > > >>>>>
> > > >>>>> Transport Area Director(s):
> > > >>>>> - Magnus Westerlund <[e-mail]>
> > > >>>>> - Lars Eggert <[e-mail]>
> > > >>>>>
> > > >>>>> Transport Area Advisor:
> > > >>>>> - Lars Eggert <[e-mail]>
> > > >>>>>
> > > >>>>> Mailing Lists:
> > > >>>>> General Discussion: [e-mail]
> > > >>>>> To Subscribe: [e-mail]
> > > >>>>> In Body: (un)subscribe
> > > >>>>> Archive: 
> http://www.ietf.org/mail-archive/web/storm/index.html
> > > >>>>>
> > > >>>>> Description of Working Group:
> > > >>>>>
> > > >>>>> The IETF IPS (IP Storage) and RDDP (Remote Direct Data
> > > >> Placement)
> > > >>>>> working groups have produced a significant number of
> > > >>>> storage protocols
> > > >>>>> (e.g., iSCSI, iSER and FCIP) for which there is
significant
> > > >>>> usage. The
> > > >>>>> time has come to reflect feedback from implementation and
> > > >>> usage into
> > > >>>>> updated RFCs; this work may include:
> > > >>>>>
> > > >>>>> - Implementation-driven revisions and updates to existing
> > > >>> protocols
> > > >>>>> (i.e., updated RFCs that match the "running code").
> > > >>>>>
> > > >>>>> - Interoperability reports as needed for the resulting
> > > >>>>> revised protocols
> > > >>>>> that are appropriate for Draft Standard RFC status.
> > > >>>>>
> > > >>>>> - Minor protocol changes or additions. Backwards
> > > >> compatibility is
> > > >>>>> required.
> > > >>>>>
> > > >>>>> Significant changes to the existing protocol standards are
> > > >>>>> out of scope,
> > > >>>>> including any work on version 2 of any of these protocols.
> > > >>>>>
> > > >>>>> Stability is critical to the usage of these protocols, so
> > > >>> backwards
> > > >>>>> compatibility with existing implementations will be a
> > > >> requirement
> > > >>>>> imposed on for all protocol changes and additions. Note
> > > >>>> that this is a
> > > >>>>> requirement for implementation compatibility - if it is
the
> > > >>>> case that
> > > >>>>> all implementations of a protocol have done something
> > > >>> different than
> > > >>>>> what the RFC specifies, it is appropriate for a new RFC to
> > > >>>>> document what
> > > >>>>> the "running code" actually does and deprecate the
> > > >> unused original
> > > >>>>> behavior.
> > > >>>>>
> > > >>>>> Initial list of work items:
> > > >>>>>
> > > >>>>> (1) iSCSI: Combine RFCs 3720 (iSCSI), 3980 (NAA names),
> > > >> 4850 (node
> > > >>>>> architecture key) and 5048 (corrections/clarifications)
> > > >>>> into one draft
> > > >>>>> (3720bis), removing features that are not implemented in
> > > >>>>> practice. This
> > > >>>>> draft should be prepared so that it could become a Draft
> > > >>>> Standard RFC,
> > > >>>>> but it is up to the to decide whether to advance it to
> > > >>>> Draft Standard.
> > > >>>>>
> > > >>>>> (2) iSCSI: Add features to support SAM-4 (4th version
> > > >> of the SCSI
> > > >>>>> architecture) in a backwards-compatible fashion, as iSCSI
> > > >>>> is currently
> > > >>>>> based on SAM-2. This will be a separate draft from the
> > > >>>> iSCSI update in
> > > >>>>> the previous bullet. The Working group may add additional
> > > >>>> minor useful
> > > >>>>> iSCSI features to this draft.
> > > >>>>>
> > > >>>>> (3) FCIP: IP Protocol number 133 was allocated to a
> > > >>> precursor of the
> > > >>>>> FCIP protocol in 2000, but this allocated number is not
> > > >>>> used by FCIP.
> > > >>>>> The working group will consider whether this 
> allocated number
> > > >>>>> should be
> > > >>>>> returned to IANA for future reallocation.
> > > >>>>>
> > > >>>>> (4) iFCP: The Address Translation mode of iFCP needs to be
> > > >>>> deprecated
> > > >>>>> (SHOULD NOT implement or use), as there are significant
> > > >> technical
> > > >>>>> problems with its specification, and moreover, only the
> > Address
> > > >>>>> Transparent mode of iFCP is in use. This will be done via
a
> > > >>>>> short draft
> > > >>>>> that updates RFC 4172, and not via a complete rewrite of
> > > >>> RFC 4172. A
> > > >>>>> combined draft is expected that encompasses items 
> (3) and (4).
> > > >>>>>
> > > >>>>> (5) RDDP MPA: Good support for MPI applications requires a
> > > >>>>> small update
> > > >>>>> to the startup functionality to allow either end of the
> > > >>>> connection to
> > > >>>>> initiate.
> > > >>>>>
> > > >>>>> (6) iSER: Experience with Infiniband implementations
suggest
> > > >>>>> a few minor
> > > >>>>> updates to reflect what has been done in practice.
> > > >>>>>
> > > >>>>> The working group is expected to maintain good working
> > > >>> relationships
> > > >>>>> with INCITS Technical Committee T10 (SCSI standards) and
> > INCITS
> > > >>>>> Technical Committee T11 (Fibre Channel standards) via
> > > >> overlaps in
> > > >>>>> membership as opposed to appointment of formal liaisons.
> > > >>> The liaison
> > > >>>>> process (including IAB appointment of a liaison or
> > > >>>>> liaisons) remains available for use if needed.
> > > >>>>>
> > > >>>>> Goals and Milestones:
> > > >>>>>
> > > >>>>> June 2009 First version of FCIP protocol number and iFCP
> > Address
> > > >>>>> Translation mode draft
> > > >>>>>
> > > >>>>> July 2009 First version of iSCSI SAM-4 (and other) new
> > > >>>> features draft.
> > > >>>>>
> > > >>>>> Aug 2009 First version of RDDP MPA startup change draft
> > > >>>>>
> > > >>>>> Sep 2009 Working Group Last Call on FCIP protocol
> > > >> number and iFCP
> > > >>>>> address change draft
> > > >>>>>
> > > >>>>> Sep 2009 First version of combined iSCSI draft (3720bis)
> > > >>>>>
> > > >>>>> Oct 2009 First version of iSER update draft
> > > >>>>>
> > > >>>>> Oct 2009 Working Group Last Call on RDDP MPA startup
> > > >> change draft.
> > > >>>>>
> > > >>>>> Dec 2009 Functionally complete iSCSI SAM-4 (and other)
> > > >>> new features
> > > >>>>> draft.
> > > >>>>>
> > > >>>>> Feb 2010 Working Group Last Call on iSER update draft
> > > >>>>>
> > > >>>>> March 2010 Working Group Last Call on iSCSI SAM-4 (and
> > > >> other) new
> > > >>>>> features draft.
> > > >>>>>
> > > >>>>> April 2010 Working Group decision on whether to seek Draft
> > > >>>>> Standard RFC
> > > >>>>> status for the combined iSCSI draft (3720bis). [Note:
> > > >>>>> decision may be made significantly before this date.]
> > > >>>>>
> > > >>>>> Sep 2010 Working Group Last Call on combined iSCSI
> > > >> draft (3720bis)
> > > >>>>> _______________________________________________
> > > >>>>> OPS-DIR mailing list
> > > >>>>> [e-mail]
> > > >>>>> https://www.ietf.org/mailman/listinfo/ops-dir
> > > >>>>>
> > > >>>>
> > > >>>
> > > >>>
> > > >>
> > > 
> > > _______________________________________________
> > > OPS-DIR mailing list
> > > [e-mail]
> > > https://www.ietf.org/mailman/listinfo/ops-dir
> > > 
> > 
> > 
> > 
>