Internet Small Computer System Interface (iSCSI) Extensions for the Remote Direct Memory Access (RDMA) Specification
draft-ietf-storm-iser-15

[Ballot comment]
In 1.1, you seem to have a diction problem in Completion:

      Completion is defined
      as the process by the RDMA-Capable Protocol layer to inform

Do you mean "Completion is defined as the process by which the RDMA-Capable Protocol layer informs"?

I assume that iSCSI control-type PDU and iSCSI data-type PDU are terms of art, and that you can't change them, but these are horribly confusing terms to use.  If possible, it would be better to leave out the "-type" part of the name, and use a name that makes clear that these are messages, and not pieces of hardware.

      iSER-IRD - This variable represents the maximum number of incoming
      outstanding RDMA Read Requests that the iSER Layer at the
      initiator declares on a particular RCaP Stream.

It might be better to say "specifies for" rather than "declares on" here.  I found the way iSER-ORD was defined to be clearer, also—this definition doesn't say what effect the iSER-IRD has.

Phase Collapse is a really confusing term—it sounds like something that happens in an inductor in an AC circuit.  Can't you say something like "Phase aggregation" instead?

1.2 does not define the acronym "RCaP".  Personally I'd prefer to see fewer acronyms anyway, but it seems inconsistent to define so many acronyms in 1.2 and not define this one.

In 3.3:
  4.  The iSCSI Layer at the initiator MUST NOT issue SNACKs for PDUs.

Which type of SNACK do you mean here?  Given that the acronym is defined as having two meanings, I think you need to specify.

[Ballot comment]
I have only reviewed the things in this document that have changed since 5046.

Overall comment: The writeup indicates that there is widespread interoperable deployment. The changes to this document from 5046 seem to document what's actually going on in the field, as against introducing new functionality. Is there a reason that you didn't submit this for full Internet Standard instead of recycling at Proposed Standard? Unless you see an update to this coming soon, it seems like a waste not to try to advance it unless there is a good reason not to.

Only a couple of specific comments:


2.5.1:

  The iSER layer at the initiator SHOULD invalidate the Advertised
  STag upon a normal completion of the associated task.  The Send with
  Invalidate Message, if supported by the RCaP layer (e.g., iWARP),
  can be used for automatic invalidation when it is used to carry the
  SCSI Response PDU.  There are two exceptions to this automatic
  invalidation - bidirectional commands, and abnormal completion of a
  command.  The iSER Layer at the initiator SHOULD explicitly
  invalidate the STag in these two cases.
 
 
The last paragraph of this section describes the downside of violating the SHOULDs in the above paragraph. Thanks for that; it's exactly what a document ought to provide whenever there is a SHOULD. But you don't explain the sorts of circumstances that would necessitate not following the SHOULDs: Why are these not MUSTs?

5.1.1:

  If the outcome of the iSCSI negotiation is to enable iSER-assisted
  mode, then on the initiator side, prior to sending the Login Request
  with the T (Transit) bit set to 1 and the NSG (Next Stage) field set
  to FullFeaturePhase, the iSCSI Layer SHOULD request the iSER Layer
  to allocate the connection resources necessary to support RCaP by
  invoking the Allocate_Connection_Resources Operational Primitive.

I don't understand why the MUST was changed to a SHOULD, but then again I don't understand why either one is used in the first place: Don't you just mean "the iSCSI Layer will request the iSER Layer to allocate" or "the iSCSI Layer can request the iSER Layer to allocate"? Maybe I'm not understanding the meaning of the requirement here. Is this part of what's being discussed in the new text in 5.1.3? If this is just about the timing, switching the MUST to a SHOULD in this paragraph doesn't make that clear.

  These may include, but not
  limited to, MaxOutstandingR2T, ErrorRecoveryLevel, etc.

5046 was correct here: "but *are* not limited to"

One more global comment, strictly editorial (and probably pedantic): You've (incorrectly) changed a bunch of "that"s to "which"s. The RFC Editor will probably deal with them, but I don't understand why you did.

[Ballot comment]
Thank you for your help in resolving my DISCUSS.

In 2.1  Motivation

  Out-of-order TCP segments in the Traditional iSCSI model have to be
  stored and reassembled before the iSCSI protocol layer within an end
  node can place the data in the iSCSI buffers.  This reassembly is
  required because not every TCP segment is likely to contain an iSCSI
  header to enable its placement and TCP itself does not have a built-
  in mechanism for signaling ULP message boundaries to aid placement
  of out-of-order segments. 

Isn't the fundamental reason out-of-order TCP segments have to be reassembled, that TCP offers a connection-oriented service with no alternatives to in-order delivery to an application? Or are you talking about something else? I note that pretty much the same text is in RFC 5046, so, I'm probably just not understanding what's going on.

In 4.1  Interactions with the RCaP Layer

  The iSER protocol layer is layered on top of an RCaP layer (see
  Error! Reference source not found.)

I think you've already seen comments about problems with cross-references, but I'm including for completeness.

In 5.1  iSCSI/iSER Connection Setup

  iSER-assisted mode MUST be enabled only if it is negotiated on the
  leading connection during the LoginOperationalNegotiation Stage of
  the iSCSI Login Phase. 

I think I understand this, but it might be clearer to say

  iSER-assisted mode MUST NOT be enabled unless it is negotiated on the
  leading connection during the LoginOperationalNegotiation Stage of
  the iSCSI Login Phase.

In 5.2.3.2  Connection Termination Notification to the iSCSI Layer

  If the remote iSCSI/iSER node initiated the closing of the
  Connection (e.g., by sending a TCP FIN or TCP RST), the iSER Layer
  MUST notify the iSCSI Layer after the RCaP layer reports that the
  Connection is closed by invoking the Connection_Terminate_Notify
  Operational Primitive.

Do the iSER and iSCSI layers treat TCP FIN and TCP RST the same way? is there any action that's required for one, but not the other? I think (from reading the sections on recovery) that they're treated the same way, so I'm not asking for a text change here, just for some clue.

In 7.3.11 SNACK Request

  Since HeaderDigest and DataDigest must be negotiated to "None",
  there are no digest errors when the connection is in iSER-assisted
  mode.  Also since RCaP delivers all messages in the order they were
  sent, there are no sequence errors when the connection is in iSER-
  assisted mode.  Therefore the iSCSI Layer MUST NOT send SNACK
  Request PDUs.  A SNCAK Request PDU, if used, MUST be treated as an
                    ^
Is this a typo?

  iSCSI protocol error.  The iSER Layer MAY reject such a PDU from the
  iSCSI Layer with an appropriate error code.  If a SNACK Request PDU
  is received by the iSCSI Layer at the target, it MUST respond with a
  Reject PDU with a reason code of "protocol error".

[Ballot discuss]
I'm a soft DISCUSS here, and this should be easy to resolve.

This specification is mostly crisp, using MUSTs, but in 2.5.2  Send, there are a lot of SHOULDs. Some are OK, some, I'm wondering whether they ought to be MUSTs, and some, I'm not sure are 2119 SHOULDs. Please catch me up?

  For interoperability, iSER implementations SHOULD accept and
  correctly process SendSE and SendInvSE messages.  However, SendSE
  and SendInvSE messages are to be regarded as optimizations or
  enhancements to the basic Send message, and their support may vary
  by RCaP protocol and specific implementation.  In general, these
  messages SHOULD NOT be used, unless the RCaP requires support for
  them in all implementations.  If these messages are used, the
  implementation SHOULD be capable of reverting to use of Send in
  order to work with a receiver that does not support these message.
  Attempted use of these messages with a peer that does not support
  them may result in a fatal error that closes the RCaP connection.
  For example, these messages SHOULD NOT be used with the InfiniBand
                              ^
I'm not understanding why this statement is needed, but not a MUST NOT.

  RCaP because InfiniBand does not require support for them in all
  cases.  New iSER implementations SHOULD use Send (and not SendSE or
                                    ^
I would feel better about this SHOULD if you gave an example of a compelling reason for doing otherwise.

  SendInvSE) unless there are compelling reasons for doing otherwise.
  Similarly, iSER implementations SHOULD NOT rely on events triggered
                                  ^
Obviously, I'm not an expert here, but is it obvious how you recover if you DO rely on these events and they aren't supported?

  by SendSE and SendInvSE, as these messages may not be used.

--

In the same way, in 10.2.1 PDU Recovery

  The iSCSI Layer at the initiator SHOULD disable iSCSI timeout-driven
  PDU retransmissions.

I'm wondering if this is a 2119 SHOULD. Is there a good reason not to disable iSCSI PDU retransmissions? If you don't disable them, is there anything you need to do differently because you're also seeing these retransmissions?

  by SendSE and SendInvSE, as these messages may not be used.

[Ballot discuss]
I'm a soft DISCUSS here, and this should be easy to resolve.

This specification is mostly crisp, using MUSTs, but in 2.5.2  Send, there are a lot of SHOULDs. Some are OK, some, I'm wondering whether they ought to be MUSTs, and some, I'm not sure are 2119 SHOULDs. Please catch me up?

  For interoperability, iSER implementations SHOULD accept and
                                              ^
I'm not sure I understand why this SHOULD is needed.

  correctly process SendSE and SendInvSE messages.  However, SendSE
  and SendInvSE messages are to be regarded as optimizations or
  enhancements to the basic Send message, and their support may vary
  by RCaP protocol and specific implementation.  In general, these
  messages SHOULD NOT be used, unless the RCaP requires support for
            ^
I'm OK with this SHOULD NOT

  them in all implementations.  If these messages are used, the
  implementation SHOULD be capable of reverting to use of Send in
                  ^
I'm OK with this SHOULD

  order to work with a receiver that does not support these message.
  Attempted use of these messages with a peer that does not support
  them may result in a fatal error that closes the RCaP connection.
  For example, these messages SHOULD NOT be used with the InfiniBand
                              ^
I'm not sure why this is needed, but not a MUST NOT.

  RCaP because InfiniBand does not require support for them in all
  cases.  New iSER implementations SHOULD use Send (and not SendSE or
                                    ^
I would feel better about this SHOULD if you gave an example of a compelling reason for doing otherwise.

  SendInvSE) unless there are compelling reasons for doing otherwise.
  Similarly, iSER implementations SHOULD NOT rely on events triggered
                                  ^
Obviously, I'm not an expert here, but is it obvious how you recover if you DO rely on these events and they aren't supported?

  by SendSE and SendInvSE, as these messages may not be used.

--

In the same way, in 10.2.1 PDU Recovery

  The iSCSI Layer at the initiator SHOULD disable iSCSI timeout-driven
  PDU retransmissions.

I'm wondering if this is a 2119 SHOULD. Is there a good reason not to disable iSCSI PDU retransmissions? If you don't disable them, is there anything you need to do differently because you're also seeing these retransmissions?

  by SendSE and SendInvSE, as these messages may not be used.

[Ballot discuss]
I'm a soft DISCUSS here, and this should be easy to resolve.

This specification is mostly crisp, using MUSTs, but in 2.5.2  Send, there are a lot of SHOULDs. Some are OK, some, I'm wondering whether they ought to be MUSTs, and some, I'm not sure are 2119 SHOULDs. Please catch me up?

  For interoperability, iSER implementations SHOULD accept and
                                                                              ^
I'm not sure I understand why this SHOULD is needed (I'm not sure why this statement is needed at all).

  correctly process SendSE and SendInvSE messages.  However, SendSE
  and SendInvSE messages are to be regarded as optimizations or
  enhancements to the basic Send message, and their support may vary
  by RCaP protocol and specific implementation.  In general, these
  messages SHOULD NOT be used, unless the RCaP requires support for
                      ^
I'm OK with this SHOULD NOT

  them in all implementations.  If these messages are used, the
  implementation SHOULD be capable of reverting to use of Send in
                              ^
I'm OK with this SHOULD

  order to work with a receiver that does not support these message.
  Attempted use of these messages with a peer that does not support
  them may result in a fatal error that closes the RCaP connection.
  For example, these messages SHOULD NOT be used with the InfiniBand
                                                        ^
I'm not sure why this is needed, but doesn't need to be a MUST NOT.

  RCaP because InfiniBand does not require support for them in all
  cases.  New iSER implementations SHOULD use Send (and not SendSE or
                                                                  ^
I would feel better about this SHOULD if you gave an example of a compelling reason for doing otherwise.

  SendInvSE) unless there are compelling reasons for doing otherwise.
  Similarly, iSER implementations SHOULD NOT rely on events triggered
                                                            ^
Obviously, I'm not an expert here, but is it obvious how you recover if you DO rely on these events and they aren't supported? If this was a MUST, I wouldn't be asking :-)

--

In the same way, in 10.2.1 PDU Recovery

  The iSCSI Layer at the initiator SHOULD disable iSCSI timeout-driven
  PDU retransmissions.

I'm wondering if this is a 2119 SHOULD. Is there a good reason not to disable iSCSI PDU retransmissions? If you don't disable them, is there anything you need to do differently because you're also seeing these retransmissions?

  by SendSE and SendInvSE, as these messages may not be used.

[Ballot comment]
In 2.1  Motivation

  Out-of-order TCP segments in the Traditional iSCSI model have to be
  stored and reassembled before the iSCSI protocol layer within an end
  node can place the data in the iSCSI buffers.  This reassembly is
  required because not every TCP segment is likely to contain an iSCSI
  header to enable its placement and TCP itself does not have a built-
  in mechanism for signaling ULP message boundaries to aid placement
  of out-of-order segments. 

Isn't the fundamental reason out-of-order TCP segments have to be reassembled, that TCP offers a connection-oriented service with no alternatives to in-order delivery to an application? Or are you talking about something else? I note that pretty much the same text is in RFC 5046, so, I'm probably just not understanding what's going on.

In 4.1  Interactions with the RCaP Layer

  The iSER protocol layer is layered on top of an RCaP layer (see
  Error! Reference source not found.)

I think you've already seen comments about problems with cross-references, but I'm including for completeness.

In 5.1  iSCSI/iSER Connection Setup

  iSER-assisted mode MUST be enabled only if it is negotiated on the
  leading connection during the LoginOperationalNegotiation Stage of
  the iSCSI Login Phase. 

I think I understand this, but it might be clearer to say

  iSER-assisted mode MUST NOT be enabled unless it is negotiated on the
  leading connection during the LoginOperationalNegotiation Stage of
  the iSCSI Login Phase.

In 5.2.3.2  Connection Termination Notification to the iSCSI Layer

  If the remote iSCSI/iSER node initiated the closing of the
  Connection (e.g., by sending a TCP FIN or TCP RST), the iSER Layer
  MUST notify the iSCSI Layer after the RCaP layer reports that the
  Connection is closed by invoking the Connection_Terminate_Notify
  Operational Primitive.

Do the iSER and iSCSI layers treat TCP FIN and TCP RST the same way? is there any action that's required for one, but not the other? I think (from reading the sections on recovery) that they're treated the same way, so I'm not asking for a text change here, just for some clue.

In 7.3.11 SNACK Request

  Since HeaderDigest and DataDigest must be negotiated to "None",
  there are no digest errors when the connection is in iSER-assisted
  mode.  Also since RCaP delivers all messages in the order they were
  sent, there are no sequence errors when the connection is in iSER-
  assisted mode.  Therefore the iSCSI Layer MUST NOT send SNACK
  Request PDUs.  A SNCAK Request PDU, if used, MUST be treated as an
                    ^
Is this a typo?

  iSCSI protocol error.  The iSER Layer MAY reject such a PDU from the
  iSCSI Layer with an appropriate error code.  If a SNACK Request PDU
  is received by the iSCSI Layer at the target, it MUST respond with a
  Reject PDU with a reason code of "protocol error".

in tracker
IANA Actions - YES

NOTE: This revised review is based on the current version 14 of
the drafted document. 

We have a question regarding one of the actions in this document.

We understand that, upon approval of this document, there are two
*additional* IANA actions:

- the following entries in the iSCSI Login/Text Keys sub-registry
should be updated to this document:

OLD:
InitiatorRecvDataSegmentLength[RFC5046]
MaxOutstandingUnexpectedPDUs[RFC5046]
RDMAExtensions[RFC5046]
TargetRecvDataSegmentLength[RFC5046]

NEW:
InitiatorRecvDataSegmentLength [RFC-to-be]
MaxOutstandingUnexpectedPDUs [RFC-to-be]
RDMAExtensions  [RFC-to-be]
TargetRecvDataSegmentLength  [RFC-to-be]

QUESTION: Please clarify if this document confirms that Standards
Action is remained as the registration procedure for the sub-registry
"iSCSI Login/Text Keys".


Please see: www.iana.org/assignments/iscsi-parameters

- the following values in the sub-registry 'iSER Opcodes' should be
updated to this document:

OLD:
0x1 iSCSI control-type [RFC5046]
0x2 iSER Hello [RFC5046]
0x3 iSER HelloReply [RFC5046]

NEW:
0x1 iSCSI control-type [RFC-to-be]
0x2 iSER Hello  [RFC-to-be]
0x3 iSER HelloReply [RFC-to-be]

www.iana.org/assignments/iscsi-parameters


Note: The actions requested in this document will not be completed
until the document has been approved for publication as an RFC.
This message is only to confirm what actions will be performed.

IANA has reviewed draft-ietf-storm-iser-12 and has the following comments:

IANA understands that, upon approval of this document, there is a single action
which IANA must complete.

In the iSCSI Login/Text Keys subregistry of the Internet Small Computer
System Interface (iSCSI) Parameters registry located at:

www.iana.org/assignments/iscsi-parameters/iscsi-parameters.xml

three new registrations will be made as follows:

Key: MaxAHSLength
Reference: [ RFC-to-be ]

Key: TaggedBufferForSolicitedDataOnly
Reference: [ RFC-to-be ]

Key: iSERHelloRequired
Reference: [ RFC-to-be ]

IANA understands that this is the only action required to be completed
upon approval of this document.

Note: The actions requested in this document will not be completed
until the document has been approved for publication as an RFC.

The following Last Call announcement was sent out:

From: The IESG
To: IETF-Announce
CC:
Reply-To: ietf@ietf.org
Subject: Last Call:  (iSCSI Extensions for RDMA Specification) to Proposed Standard


The IESG has received a request from the STORage Maintenance WG (storm)
to consider the following document:
- 'iSCSI Extensions for RDMA Specification'
  as Proposed Standard

The IESG plans to make a decision in the next few weeks, and solicits
final comments on this action. Please send substantive comments to the
ietf@ietf.org mailing lists by 2012-10-22. Exceptionally, comments may be
sent to iesg@ietf.org instead. In either case, please retain the
beginning of the Subject line to allow automated sorting.

Abstract


  iSCSI Extensions for RDMA provides the RDMA data transfer capability
  to iSCSI by layering iSCSI on top of an RDMA-Capable Protocol.  An
  RDMA-Capable Protocol provides RDMA Read and Write services, which
  enable data to be transferred directly into SCSI I/O Buffers without
  intermediate data copies.  This document describes the extensions to
  the iSCSI protocol to support RDMA services as provided by an RDMA-
  Capable Protocol.

  This document obsoletes RFC 5046.




The file can be obtained via
http://datatracker.ietf.org/doc/draft-ietf-storm-iser/

IESG discussion can be tracked via
http://datatracker.ietf.org/doc/draft-ietf-storm-iser/ballot/


No IPR declarations have been submitted directly on this I-D.

AD review:

INTRODUCTION, paragraph 11:

> iSCSI Extensions for RDMA provides the RDMA data transfer capability
> to iSCSI by layering iSCSI on top of an RDMA-Capable Protocol.  An
> RDMA-Capable Protocol provides RDMA Read and Write services, which
> enable data to be transferred directly into SCSI I/O Buffers without
> intermediate data copies.  This document describes the extensions to
> the iSCSI protocol to support RDMA services as provided by an RDMA-
> Capable Protocol.

  Expand RDMA on first use in the Abstract, but also later on.


Section 1.1, paragraph 1:

> Advertisement (Advertised, Advertise, Advertisements, Advertises) -
> The act of informing a remote iSER Layer that a local node's buffer
> is available to it.  A Node makes a buffer available for incoming
> RDMA Read Request Message or incoming RDMA Write Message access by
> informing the remote iSER Layer of the Tagged Buffer identifiers
> (STag, Base Offset, and buffer length).  Note that this Advertisement
> of Tagged Buffer information is the responsibility of the iSER Layer
> on either end and is not defined by the RDMA-Capable Protocol.  A
> typical method would be for the iSER Layer to embed the Tagged
> Buffer's STag, Base Offset, and buffer length in a message destined
> for the remote iSER Layer.

  Expand iSER on first use.


Section 1.1, paragraph 6:

> Data Sink - The peer receiving a data payload.  Note that the Data
> Sink can be required to both send and receive RCaP Messages to
> transfer a data payload.

  Expand RCaP on first use, though it is introduced later on. This
  will improve the readability.


Section 1.1, paragraph 38:

> RDMA Protocol (RDMAP) - A wire protocol that supports RDMA Operations
> to transfer ULP data between a Local Peer and the Remote Peer as
> described in [RDMAP].

  Expand ULP on first use.

Section 2.7 in general
I would like to see this section more upfront in Section 2, probably at
the beginning. The reason for this is, that I was looking all the time
for an overview picture to get all the involved levels arranged in my
head. Section 2.7 helped a lot, but it came too late for me.

Section 12, paragraph 6:

> IANA is requested to update the registrations of the other 4 iSER
> keys in that registry to reference the RFC number of this draft when
> it is published as an RFC.

  You are referring to the 'iSER Opcodes' table, isn't it? Please
  state this in the text, just to avoid any confusion.

PROTO writeup:

iSCSI Extensions for RDMA Specification
draft-ietf-storm-iser-11.txt

PROTO shepherd: David L. Black (STORM WG Co-Chair)
------------------------------------------------------------------------

(1) What type of RFC is being requested (BCP, Proposed Standard,
Internet Standard, Informational, Experimental, or Historic)? Why
is this the proper type of RFC? Is this type of RFC indicated in the
title page header?

Proposed Standard RFC is requested because this draft updates and
replaces RFC 5046, a Proposed Standard RFC. Proposed Standard is
indicated as the intended status is in the title page header.

(2) The IESG approval announcement includes a Document Announcement
Write-Up. Please provide such a Document Announcement Write-Up. Recent
examples can be found in the "Action" announcements for approved
documents. The approval announcement contains the following sections:

Technical Summary

iSCSI Extensions for RDMA provides the RDMA data transfer capability
to iSCSI by layering iSCSI on top of an RDMA-Capable Protocol. An
RDMA-Capable Protocol provides RDMA Read and Write services, which
enable data to be transferred directly into SCSI I/O Buffers without
intermediate data copies. This document describes the extensions to
the iSCSI protocol to support RDMA services as provided by an RDMA-
Capable Protocol.

Working Group Summary

This document is a minor update to RFC 5046, primarily to reflect
what has actually been done in implementations. WG Last Call turned
up a controversial issue around relaxing RFC 5046's requirements for
use of versions of the Send message that include STage invalidation
and posting of a solicited event as some implementations have ignored
those requirements. The current version of this document resolves
that issue.

Document Quality

There are multiple implementations of the iSER protocol;
the primary purpose of this document is to reflect implementation
experience so that the iSER protocol specification matches the
"running code". Hemal Shah's review of the document resulted in
some important changes in the text describing use of versions of
the Send message.

Personnel

Document Shepherd: David Black (storm WG co-chair)
Responsible Area Director: Martin Stiemerling (Transport)

(3) Briefly describe the review of this document that was performed by
the Document Shepherd. If this version of the document is not ready
for publication, please explain why the document is being forwarded to
the IESG.

The Document Shepherd read the document in its entirety for WG Last
Call and compared it to RFC 5046 at that timme. The Document Shepherd
has reviewed all of the subsequent changes. Numerous changes have
been made to this document in response to the Document Shepherd's
comments at WG Last Call and subsequently. The Document Shepherd
believes that this document is now ready for RFC publication.

(4) Does the document Shepherd have any concerns about the depth or
breadth of the reviews that have been performed?

No.

(5) Do portions of the document need review from a particular or from
broader perspective, e.g., security, operational complexity, AAA, DNS,
DHCP, XML, or internationalization?

No.

(6) Describe any specific concerns or issues that the Document Shepherd
has with this document that the Responsible Area Director and/or the
IESG should be aware of? For example, perhaps he or she is uncomfortable
with certain parts of the document, or has concerns whether there really
is a need for it. In any event, if the WG has discussed those issues and
has indicated that it still wishes to advance the document, detail those
concerns here.

None.

(7) Has each author confirmed that any and all appropriate IPR
disclosures required for full conformance with the provisions of BCP 78
and BCP 79 have already been filed. If not, explain why.

Yes.

(8) Has an IPR disclosure been filed that references this document?

No.

(9) How solid is the WG consensus behind this document? Does it
represent the strong concurrence of a few individuals, with others
being silent, or does the WG as a whole understand and agree with it?

The storm (STORage Maintenance) WG is a maintenance WG that works on
a number of storage technologies, and hence not every participant is
interested in every technology. The members of the WG who are
interested in iSER understand and agree with this document.

(10) Has anyone threatened an appeal or otherwise indicated extreme
discontent?

No.

(11) Identify any ID nits the Document Shepherd has found in this
document. (See http://www.ietf.org/tools/idnits/ and the Internet-Drafts
Checklist). Boilerplate checks are not enough; this check needs to be
thorough.

No nits - idnits 2.12.13 issues three warnings that do not reflect
actual problems with the draft.

(12) Describe how the document meets any required formal review
criteria, such as the MIB Doctor, media type, and URI type reviews.

Not applicable.

(13) Have all references within this document been identified as
either normative or informative?

Yes.

(14) Are there normative references to documents that are not ready for
advancement or are otherwise in an unclear state?

No.

(15) Are there downward normative references references (see RFC 3967)?

No.

(16) Will publication of this document change the status of any
existing RFCs? Are those RFCs listed on the title page header, listed
in the abstract, and discussed in the introduction? If the RFCs are not
listed in the Abstract and Introduction, explain why, and point to the
part of the document where the relationship of this document to the
other RFCs is discussed. If this information is not in the document,
explain why the WG considers it unnecessary.

This document obsoletes RFC 5046; that is listed in the title page header
and stated in the Abstract. The Introduction section starts on p.15, and
it does not seem necessary to repeat the obsolescence statement that far
into the document.

(17) Describe the Document Shepherd's review of the IANA considerations
section, especially with regard to its consistency with the body of the
document. Confirm that all protocol extensions that the document makes
are associated with the appropriate reservations in IANA registries.
Confirm that any referenced IANA registries have been clearly identified.

The IANA considerations section adds three keys to the iSCSI
Login/Text Keys" registry of "iSCSI Parameters", and requests reference
updates to other iSER keys that will be confirmed by IANA. The document
shepherd has checked the IANA Considerations section and believes it
to be correct and sufficient.

(17-continued) Confirm that newly created IANA registries include a
detailed specification of the initial contents for the registry, that
allocations procedures for future registrations are defined, and a
reasonable name for the new registry has been suggested (see RFC 5226).

Not applicable.

(18) List any new IANA registries that require Expert Review for future
allocations. Provide any public guidance that the IESG would find
useful in selecting the IANA Experts for these new registries.

Not applicable.

(19) Describe reviews and automated checks performed by the Document
Shepherd to validate sections of the document written in a formal
language, such as XML code, BNF rules, MIB definitions, etc.

Not applicable.