[Gen-art] Review: draft-ietf-ipfix-rfc5815bis-01.txt

["Joel M. Halpern" <[e-mail]>]

I am the assigned Gen-ART reviewer for this draft. For background on 
Gen-ART, please see the FAQ at 
<http://wiki.tools.ietf.org/area/gen/trac/wiki/GenArtfaq>.

Please resolve these comments along with any other Last Call comments 
you may receive.

Document: draft-ietf-ipfix-rfc5815bis-01.txt
     Definitions of Managed Objects for IP Flow Information Export
Reviewer: Joel M. Halpern
Review Date: 23-Feb-2012
IETF LC End Date: 8-March-2012
IESG Telechat date: 15-March-2012

Summary: This document is nearly ready for publication as a Proposed 
Standard.  While each individual item below is minor, the collection is 
sufficiently confusing that I would like to see them resolved before 
publication.

Major issues:

Minor issues:
     The description in the first paragraph of 5.2 on the Template table 
is written in such a way as to lead the reader to think the Observation 
Domain is somehow associated with the Transport Session (which table 
provides the device mode which is discussed in that text.)  Could you 
split the Observation Domain reference out to a separate sentence 
(possibly before the reference to the Transport Session)?

     In the example of the export table in section 5.4, there are two 
blocks of export entries, on with ipfixExportIndex 7, and one with 8. 
They are mirrors of each other except that they reverse which transport 
session is primary, and which is secondary.  There is no explanation of 
why both are present.  And there is no explanation of why the settings 
in index 7 are used, rather than those in index 8.

     If you define an Observation Point with 0 for both Physical Entity 
and Physical Interface, what is it observing?  Similarly, what is a 
metering process metering if its Observation Point Group Ref is 0?