Document: draft-ietf-ips-iser-06.txt
          iSCSI Extensions for RDMA Specification
Reviewer: Joel Halpern
Review Date:  January 12, 2007
IETF LC Date: January 3, 2007

Summary: This document is ready for publication as a Proposed Standard

Comments:

Nit:  The references to "remote iSER Layer" as a participant in 
communication (in the definitions section) is odd.  I've seen 
references to Layer Instances, Layer Entities, and other 
qualifiers.  But It is unusual to see Layer by itself used this 
way.  (Normally, a layer refers to the collection of communicating 
entities in that layer.  Even the definitions in this document define 
it that way.  A layer is neither local nor remote.)  I suspect that 
most of these occurrences would be better as "Remote Peer" according 
to the definitions. (or remote iSER node or more verbosely the iSER 
layer at the remote node, as per later usage.)

Nit: After the first occurrence of "RDMA-Capable Protocol" in the 
abstract, could you just put "(RCaP)"

Nit: I would include all the acronyms referenced in the definitions 
(SN, RCaP, ...) into the Acronym list.

Nit: I may have missed it, but I did not find a definitions (or 
description) of target and initiator before 2.4.  I presume that this 
is well-defined in other RDMA specs.  A paragraph here (or actual 
definitions in section 1) would be helpful.  (Yes, this can be 
understood once one gets to 2.5.  It is a matter of avoidin