Re: [OSPF] MIB Dr. Review of draft-ietf-ospf-ospfv3-mib-12.txt

["Daniel Joyal" <djoyal at nortel.com>]

Joan,

 Thanks for the prompt and very thorough review.
We will get to work addressing the comments.

Regards,
-Dan 

> -----Original Message-----
> From: Joan Cucchiara [mailto:jcucchiara at mindspring.com] 
> Sent: Friday, March 07, 2008 2:38 AM
> To: Romascanu, Dan (Dan); Acee Lindem; ospf at ietf.org; MIB 
> Doctors (E-mail)
> Cc: Ronald Bonica; Abhay Roy; Ross Callon; Vishwas Manral; 
> Joyal, Daniel (BL60:SF23); Bert Wijnen; David Ward
> Subject: MIB Dr. Review of draft-ietf-ospf-ospfv3-mib-12.txt
> 
> 
> Hi Folks,
> 
> First is the output from the MIB tools.
> Followed by some General Comments, and then followed by 
> fairly specific comments/questions on the MIB.
> 
> Thanks,
>    Joan
> 
> 
> SMICNG results:
> ----------------
> 
> E: f(OSPFV3-MIB.txt), (2523,21) Item "ospfv3AreaAggregateRouteTag"
> in sequence "Ospfv3AreaAggregateEntry" has conflicting syntax 
> specified
> 
> E: f(OSPFV3-MIB.txt), (756,31) Index item "ospfv3AsLsdbLsid"
> must be defined with syntax that includes a range
> 
> E: f(OSPFV3-MIB.txt), (904,31) Index item "ospfv3AreaLsdbLsid"
> must be defined with syntax that includes a range
> 
> E: f(OSPFV3-MIB.txt), (1066,31) Index item "ospfv3LinkLsdbLsid"
> must be defined with syntax that includes a range
> 
> E: f(OSPFV3-MIB.txt), (2647,31) Index item "ospfv3VirtLinkLsdbLsid"
> must be defined with syntax that includes a range
> 
> 
> smilint
> ------------
> severity level requested 6
> 
> line  severity  problem
> --------------------------
> 
> 2844 5 warning: `InetAddress' object should have an
>                 accompanied preceding `InetAdressType' object 
> (Please see comments below wrt Notifications)
> 
> 
> 
> ID NITS (== 1 warning)
> --------
> 
>  == The copyright year in the IETF Trust Copyright Line does 
> not match the
>      current year
> 
> 
> 
> 
> General Comments
> -----------------
> 
> 1) Need to have a read-only conformance.
> 
> 2) I would like to understand the use of the single 
> DiscontinuityTime object for all the counters in the MIB.  My 
> thoughts are that having a few more discontinuity objects 
> (specifically in tables) would be more beneficial, so please 
> give me some feedback on this, so I can understand why only 
> one Discontinuity object is used.
> 
> 3) Throttling of notifications and Polling Event Counters.
> I think more details of this need to be in the MIB itself.
> Please expand some of these DESCRIPTIONS.
> 
> 4) Please note that many of my comments apply to the entire 
> MIB, although I only commented on the first few objects.  
> Please be sure to check the entire MIB for an issue.  For 
> example, please rename any RowStatus object with 
> xxxRowStatus, please spell out Interval, etc.
> 
> 5) The MIB did not extract using mstrip (MIB tool).
> If possible, could this be fixed?
> 
> 
> Comments in Order of the document
> ----------------------------------
> 
> 1) (NIT) Header should say:
> Proposed Status: Standards Track
> 
> 
> 2) (NIT) Expiration date has incorrect year.
> 
> 
> *) (NIT) please be consistent with using RFCXXX vs. MIB tree 
> assignment, so for example
> 
>              DESCRIPTION
>                  "The MIB module for OSPF version 3.
> 
>                  Copyright (C) The IETF Trust (2007).
>                  This version of this MIB module is part of
>                  RFC XXXX;  see the RFC itself for full legal
>                  notices."
> 
>              REVISION "200709171200Z"
>              DESCRIPTION -- RFC Editor assigns RFC XXXX
>                  "Initial version, published as RFC XXXX"
> 
>              ::= { mib-2 YYY } -- to be assigned by IANA
> 
> 
> 
> 
> *)Ospfv3UpToRefreshIntervalTc
> 
>   *) Please change to use TC (in caps) so, Ospfv3UpToRefreshIntervalTC
> 
>   *)  Could  DISPLAY-HINT "d" be used instead?
> 
>   *) Should be Unsigned32 (see rfc4181.txt section 4.6.1.1.)
> 
>   *) Please add a reference
> 
> *) Ospfv3DeadIntRangeTc
>   *) Please spell out Interval and change Tc to TC,
>       i.e. Ospfv3DeadIntervalRangeTC
> 
>   *) Could  DISPLAY-HINT "d" be used instead?
> 
>   *) Should be Unsigned32 (see rfc4181.txt section 4.6.1.1.)
> 
>   *) Please add a reference
> 
> *) Ospfv3RouterIdTC
> 
>   *) Please use TC, Ospfv3RouterIdTC
> 
>   *) Could  DISPLAY-HINT "d" be used instead?
> 
>   *) Please add a reference
> 
> *) Ospfv3AreaIdTc
> 
>   *) Please use TC, Ospfv3AreaIdTC
> 
>   *) Could  DISPLAY-HINT "d" be used instead?
> 
> 
>   *) Please add a reference
> 
> 
> *) Ospfv3IfInstIdTc
> 
>   *) Please Tc to TC, i.e. Ospfv3IfInstIdTC
> 
>   *) Could  DISPLAY-HINT "d" be used instead?
> 
>   *) Should be Unsigned32 (see rfc4181.txt section 4.6.1.1.)
> 
>   *) Please add a reference
> 
> 
> *) ospfv3RouterId
> 
> Isn't this a 32-bit unsigned integer?
> Please add REFERENCE clause.
> 
> 
> *) ospfv3AreaBdrRtrStatus
> 
> Please indicate the value of the flag
> when the router is an area border router.
> (i.e. when the value of this object is true(1)...)
> 
> *) ospfv3AsScopeLsaCksumSum
> 
>   *) should be Unsigned32
> 
> 
> *) ospfv3DemandExtensions
> 
> Same comment as with the previous object that has SYNTAX of 
> TruthValue, please specify, when this object has the value of 
> true(1), then the router supports demand circuits...
> 
> *) ospfv3ReferenceBandwidth
> 
> -Please add UNITS
> -Please add DEFVAL
> - Please add REFERENCE
> 
> 
> *) ospfv3RestartSupport
> 
> -Is there a value which could be considered a DEFVAL for this object?
> (In general, please review all the read-write objects and add 
> DEFVALs if appropriate).
> 
> *) ospfv3RestartInterval
> -Is there a value which could be considered a DEFVAL for this object?
> 
> 
> *) ospfv3RestartStrictLsaChecking
> 
> Please indicate what true(1) means in the DESCRIPTION (Please 
> do this for all objects which contain a
> TruthValue)
> 
> *) ospfv3RestartExitRc
> 
> What does the Rc stand for, could this be spelled out?
> 
> Are any additional objects needed to tell when the value of 
> this object changed?
> 
> *) ospfv3NotificationEnable
> 
> Why is this object needed?  Or,in other words, why can't 
> RFC3413 be used.
> 
> *) ospfv3StubRouterAdvertisement
> 
> Need to add DEFVAL
> Please add a REFERENCE
> 
> *) ospfv3AreaTable
> 
> - This table has a combination of both objects to configure 
> areas and statistics for areas.  Could this be devided into
> 2 tables one for configuring and one for status/statics?
> 
> Additionally, if is not clear to me what happens when an 
> attached area is established and then the configuration 
> changes?  Can the configurable objects be changed?
> 
> Also, the counters in this table refer to 
> ospfv3DiscontinuityTime but if the OSPFv3 routing process 
> starts-up, then my understanding is that these counters might 
> be affected (specifically, ospfv3AreaNssTranslatorEvents), so 
> I think another discontinuityTime object is needed...please comment.
> 
> 
> *) ospfv3AreaId
> 
> Isn't this unsigned?
> 
> 
> *) ospfv3AreaBdrRtrCount and other Gauge32 objects in this
> table may have a DEFVAL clause of zero (this is mentioned in
> the DESCRIPTION clauses).
> 
> *) ospfv3AreaScopeLsaCksumSum
> 
> Should this be Unsigned32?
> 
> 
> *) ospfv3AreaStatus
> 
> Please rename to ospfv3AreaRowStatus
> 
> *) ospfv3AreaStubMetric
> 
> Is there a DEFVAL for this object?
> 
> 
> *) ospfv3AreaNssaTranslatorStabInt
> 
> Please use entire word Interval
> 
> 
> 
> *) ospfv3AsLsdbSequence
> 
> Should this have a range?
> 
> *) ospfv3AreaLsdbSequence
> 
> (same question, is there a range for this?)
> 
> *) ospfv3AreaLsdbAge
> 
> Unsigned32 and a range?
> 
> What is the value if the DoNotAge Bit is set?
> 
> *) ospfv3AreaLsdbChecksum
> 
> Should this be unsigned32?
> 
> 
> *) ospfv3AreaLsdbTypeKnown
> 
> Please indicate what true(1) means in the
> DESCRIPTION
> 
> *) ospfv3LinkLsdbSequence
> 
> Same comment as with the other Sequence numbers,
> should this have a range?  Additionally, should these
> sequence numbers have a common TC?
> 
> *)ospfv3LinkLsdbAge
> 
> same comments and questions
> as with the "Age" object in the prior table?
> 
> *) ospfv3LinkLsdbChecksum
> 
> Unsigned32?
> 
> 
> *) ospfv3HostTable
> 
> Need to have a storageType object added to this table.
> 
> 
> 
> 
> 
> *) ospfv3HostStatus
> 
> Please rename to ospfv3HostRowStatus
> Usually, this is the last in the list of objects.
> 
> *)ospfv3IfTable
> 
> Don't understand the REFERENCE here, please explain.
> 
> 
> Same comment as with the other table, there is mix of
> configuration and status/statistics in this one table.
> Could this be made into 2 tables?
> 
> Also, the Counters in this table
> likely need their own discontinuityTime object....
> 
> Can configurable objects change once the interface
> is established (for example, if the interface is pointToPoint?)
> 
> 
> 
> 
> *) ospfv3IfTransitDelay
> 
> Don't understand the DESCRIPTION here.  Could you
> please expand.
> 
> Is there a REFERENCE that could be added?
> 
> *) ospfv3IfStatus
> 
> Please rename to ospfv3IfRowStatus
> 
> *) ospfv3IfAdminStat
> 
> Would an OperStatus object be beneficial in this
> table?
> 
> *) ospfv3IfDemandNbrProbeRetxLimit
> 
> Retx is used for Retransmission, elsewhere Retrans
> was used.  Please be consistant.
> 
> *) ospfv3VirtIfTable
> 
> Don't understand the reference here.  If this is
> an IPv6/V3 specific table then why are you
> Refering to OSPFv2?  Is this correct?  Additionally,
> could a OSPFv3 REFERENCE be added also?
> 
> Same comments as with the other tables that have
> a mix of config and status/statistics objects.
> 
> Please change RowStatus objects to have RowStatus
> at the end of their name.
> 
> *) ospfv3NbrTable
> 
> Same question as to how the Counters in this table
> can successfully use ospfv3DiscontinuityTime.  In other
> words, if this table is specific to neighbors, then
> the counters are also, so there might be (at least it
> seems to me) a case where an entry in this table is
> affected and neighbor counters suffer a discontinuity
> but the ospfv3DiscontinuityTime may not.  Please comment.
> 
> 
> *) ospfv3NbrTable and ospfv3CfgNbrTable
> 
> What is the relationship between these 2 tables?
> The indexing is different between the 2 tables and so
> I'm wondering what the relationship is between them.
> 
> 
> 
> *) ospfv3AreaAggregateRouteTag
> 
> Is this Unsigned32?
> 
> 
> Notifications
> ------------
> 
> In general the accessible-for-notify objects should probably be
> changed into read-only objects.  Additionally, a timestamp should
> be associated with some of these.  There is an assumption made that
> it is better to send out notifications rather than poll, but I'd
> rather give the operator the choice on that.
> 
> 
> 
> 
> 
> 
> 
> Compliance Statements
> -------------------------
> 
> * Need a Read-only compliance statement.
> It is missing.
> 
> ---
> 
> 
> 
> 
> 
> 
> 
_______________________________________________
OSPF mailing list
OSPF at ietf.org
https://www.ietf.org/mailman/listinfo/ospf