Re: [eman] WG Last Call for draft-ietf-eman-energy-monitoring-mib-08 and draft-ietf-eman-energy-aware-mib-13

[Alan Luchuk <[e-mail]>]

Hello,

Here are comments about  draft-ietf-eman-energy-monitoring-mib-08 and 
draft-ietf-eman-energy-aware-mib-13.  Due to the holidays, I was unable
to get these sent to the WG E-mail list sooner.


Regarding draft-ietf-eman-energy-aware-mib-13:
----------------------------------------------

Section 5, Page 6:

   The UML on Page 6 _looks_ incorrect.  It _looks_ like the eoRelationTable
   is subordinate to, or a child of, the eoEntry.  I'm not sure how the
   actual relationship specified in the MIB is _supposed_ to be represented
   in UML.


Page 22:

   The eoRelationID in the eoRelationTable is "read-only".  In many cases,
   I do not see how an energy object can determine the eoRelationID to the
   Energy Objects to which it is connected, so I _think_ the eoRelationID 
   must be configured by the EnMS.  I _think_ the eoRelationID should be
   a read-write object like the eoRelationship object.

   As an example, consider a manageable, metering power distribution 
   unit (PDU), I don't see how it can autodiscover and populate the 
   eoRelationID of the energy objects to which it is connected.  The useful-
   ness of the eoRelationID MIB object is reduced if it is only populated
   in devices where it can be autodiscovered.



Regarding draft-ietf-eman-energy-monitoring-mib-08:
---------------------------------------------------

Section 5, 3rd paragraph, Page 6:

       "entity4CRCompliance.  This compliance requires that the 
        following 3 MIB objects from ENTITY-MIB [RFC6933]
        (entPhysicalIndex, entPhysicalName and entPhysicalUUID) MUST be 
        implemented."

   I think entPhysicalClass should be included in this list.  There are
   other references to the entity4CRCompliance in this draft that need the
   same update.


Section 5, Page 7:

          |   +-- r-n INTEGER           eoMeasurementCaliber(5)

   I think the above should be eoPowerMeasurementCaliber(5).

          |      +-- r-n Interger32        eoPowerStateMaxPower(2)

   Change "Interger32" to Integer.

          |   +-- r-n Integer32    eoEnergyParametersIntervalMode(5)

   I think the type of "eoEnergyParametersIntervalMode" should be INTEGER.


Section 5, Page 8:

   I suggest changing "powerAttributesMIB" to "POWER-ATTRIBUTES-MIB" for
   for consistency with previous text that references the ENERGY-OBJECT-MIB
   and POWER-ATTRIBUTES-MIB.

   A quick summary of the function of each of the tables in the POWER-
   ATTRIBUTES-MIB, similar to the quick summary of each of the tables in
   the ENERGY-OBJECT-MIB (on Page 6) might be helpful and consisent.


          |   +-- r-n Interger32 eoACPwrAttributesAvgVoltage(2)

      |   +-- r-n Interger32
          |                   eoACPwrAttributesTotalActivePower(7)

   Change "Interger32" to Integer.

   In the diagram for the eoACPwrAttributesEntry, after eoACPwrAttributes-
   ThdAmpheres, the eoACPwrAttributesThdVoltage object is missing.

   Also, here, and several places in the document, "Ampheres" is referenced.
   Not sure if "Ampheres" is intentional, but "Amperes" might be preferable.
   It might be good to do a case-insensitive search for "ampheres".


Section 5, Page 9:

          +-- eoACPwrAttributesDelPhaseEntry(1)
          |     |   [entPhysicalIndex, eoPhaseIndex]
          |     |                           
          |     +-- r-n Integer32
          |     |    eoACPwrAttributesDelPhaseToNextPhaseVoltage(1)
          |     +-- r-n Integer32
          |     | eoACPwrAttributesDelThdPhaseToNextPhaseVoltage(2)
          |     +-- r-n Integer32  
                                 eoACPwrAttributesDelThdCurrent(3)

   I think the column numbers 1, 2, and 3, are incorrect.  The actual column 
   numbers in the MIB are 2, 3, and 4.  Perhaps the MIB table should be fixed 
   so the column numbers are 1, 2, and 3.
   

Section 5, Page 12.

   At the very top, in the diagram for the eoACPwrAttributesEntry, after 
   eoACPwrAttributesThdAmpheres, the eoACPwrAttributesThdVoltage object 
   is missing.


Section 5.4, Page 15:

   I suggest changing "powerAttributesMIB" to "POWER-ATTRIBUTES-MIB" for
   for consistency with previous text that references the ENERGY-OBJECT-MIB
   and POWER-ATTRIBUTES-MIB.


Section 6, Page 20:

   This section references the EMAN-MON-MIB.  I think this should be
   ENERGY-OBJECT-MIB.


Section 9, Page 27:

   This section references the "energyObjectMIBObject".  Not sure what this
   is.


In the MIBs, adding eye-catching comments between the tables would ease
visually identifying where each MIB table begins.


Page 37, "eoPowerOperState" MIB object:

   The second paragraph of the DESCRIPTION clause reads:  "Possible values
   of eoPowerAdminState...".   I think "eoPowerAdminState" should be changed
   to "eoPowerOperState."


Page 38, "eoPowerStateTable":

   The text reads:

       This table has an expansion-dependent relationship on the
       eoPowerTable, containing rows describing each Power State
       for the corresponding Energy Object. For every Energy
       Object in the eoPowerTable, there is a corresponding
       entry in this table."

   Not sure I correctly understand this table, but should the second sentence
   read something like:  "For every Energy Object in the eoPowerTable, there 
   is a corresponding set of entries in this table, one entry for each power 
   state supported by the Energy Object."
  

Page 38, "eoPowerStateEntry":

   In the DESCRIPTION clause, the state numbers and state names do not match
   any documented enumerations for the IANAPowerStateSet textual convention.
   Should the existing values be replaced with:

           eman(1024),       -- indicates EMAN set
           emanmechoff(1025),
           emansoftoff(1026),
           emanhibernate(1027),
           emansleep(1028),
           emanstandby(1029),
           emanready(1030),
           emanlowMinus(1031),
           emanlow(1032),
           emanmediumMinus(1033),
           emanmedium(1034),
           emanhighMinus(1035),
           emanhigh(1036)


Page 45, "eoEnergyStored":

   The first sentence of the DESCRIPTION clause reads:

   "This object indicates the resultant of the energy consumed..."

   Perhaps the following text would be clearer:

   "This object indicates the difference of the energy consumed..."
                              ^^^^^^^^^^


Page 46, "eoEnergyMaxConsumed":

   The DESCRIPTION text reads:

       "This object is the maximum energy ever observed in
       eoEnergyConsumed since the monitoring started. This value
       is specified in the common billing units of watt-hours
       with the magnitude of watt-hours (kW-Hr,   MW-Hr, etc.)
       indicated separately in eoEnergyUnitMultiplier."

   Does this mean that the eoEnergyMaxConsumed value must be retained
   across reboots, and thus must be stored in non-volatile memory?


Page 48/49/54, where "entity4CRCompliance" is listed:

   I believe the "entPhysicalClass" object must be added to the lists of
   MIB objects (entPhysicalIndex,  entPhysicalName and entPhysicalUUID).


Page 56, "eoACPwrAttributesAvgCurrent":

   It might be niced to expand the DESCRIPTION clause so that it more 
   precisely specifies the significance of this, perhaps similar to the 
   DESCRIPTION clause for "eoACPwrAttributesAvgVoltage".


Page 56, "eoACPwrAttributesFrequency":

   The UNITS clause does not match the comment after the SYNTAX clause.

        eoACPwrAttributesFrequency OBJECT-TYPE
            SYNTAX          Integer32 (4500..6500) -- UNITS 0.01 Hertz
            UNITS           "hertz"


Would it be useful or helpful to have UnitsMultiplier companion objects for 
the following objects?

   Page 58, "eoACPwrAttributesThdAmpheres":
   Page 60, "eoACPwrAttributesPhaseAvgCurrent":
   Page 62, "eoACPwrAttributesDelPhaseToNextPhaseVoltage":
   Page 63, "eoACPwrAttributesWyePhaseToNeutralVoltage":
   Page 64, "eoACPwrAttributesWyePhaseCurrent":
                     

Page 59, "eoACPwrAttributesPhaseEntry":
   
   Under the DESCRIPTION clause, it might be helpful to add a comment that
   indicates the phase-to-phase voltages are in separate tables.  

   When I first encountered the "eoACPwrAttributesPhaseEntry", I wondered 
   about the voltages until I read further in the MIB and found the 
   eoACPwrAttributesDelPhaseTable and the eoACPwrAttributesWyePhaseTable.



Page 60, "eoACPwrAttributesPhaseActivePower",
         "eoACPwrAttributesPhaseReactivePower",
         "eoACPwrAttributesPhaseApparentPower":

   Expanding the DESCRIPTION clauses to note these are scaled by the
   eoACPwrAttributesPowerUnitMultiplier might be helpful, but the 
   DESCRIPTION clause for the "eoACPwrAttributesPowerUnitMultiplier"
   does note this scaling relationship.


Page 61, "eoACPwrAttributesPhaseImpedance":

   Are the UNITS of "volt-amperes" correct?   



Page 62/63, "eoACPwrAttributesDelPhaseToNextPhaseVoltage",
            "eoACPwrAttributesDelThdPhaseToNextPhaseVoltage",
            "eoACPwrAttributesDelThdCurrent":

   The last SID values for these objects are 2, 3, and 4, which distinctly
   does NOT match the values shown in the figure on Page 9, which shows
   the last SID values as 1, 2, and 3.


Page 74, Section 13:

        [EMAN-AWARE-MIB] J. Parello, B. Claise and M. Chandramoili,
                "draft-ietf-eman-energy-aware-mib-11 ", work in
                progress, November 2013.

   Does it matter, or should this reference draft 13, 
   draft-ietf-eman-energy-aware-mib-13?


Regards,
--Alan

 ------------------------------------------------------------------------------
 Alan Luchuk               SNMP Research, Inc.          Voice:  [phone]
 Senior Software Engineer  3001 Kimberlin Heights Road  FAX:    [phone]
 luchuk at snmp.com        Knoxville, TN  37920-9716    http://www.snmp.com/
 ------------------------------------------------------------------------------