RE: draft-ietf-ospf-mib-update-04

[Spencer.Giacalone@predictive.com]

Hold on there, guy. All lot of what your finding is legacy.. I don't know 
if I agree with turning everything on in your .inc file- you might want to 
think about that... 






"Chapman, Ken" <KChapman@unispherenetworks.com>
10/19/00 02:57 PM

 
        To:     "'Spencer.Giacalone@predictive.com'" <Spencer.Giacalone@predictive.com>
        cc:     djoyal@quantumbridge.com, fred@cisco.com, mibs@ops.ietf.org, 
ospf@discuss.microsoft.com, rcoltun@redback.com
        Subject:        RE: draft-ietf-ospf-mib-update-04


Spence,
I'm using SMICng (v2.2.07), and that is how I found these. 
You have to enable all the checking (except B, which I think is 
inconsistant
with RFC2578).

Now, I've found more bugs. 

Integer objects used as INDEX (auxiliary) objects can't be negative 
(RFC2578
clause 7.7 bottom of page 27). 
Therefore, you need to change the SYNTAX for ospfAddressLessIf,
ospfIfMetricAddressLessIf, ospfNbrAddressLessIndex, and
ospfLocalLsdbAddressLessIf from Integer32 to Unsigned32.
Note that ospfNbrAddressLessIndex actually uses the TEXTUAL-CONVENTION
InterfaceIndex, which conflicts with the TC in the IF-MIB with the same 
name
and therefore should not be used (besides, you really want the full 32 
bits
to accomidate IP addresses).
Therefore I recommend using Unsigned32 and removing the InterfaceIndex TC
altogether.

For consistancy (with updating to SMIv2), I suggest importing mib-2 from
SNMPv2-MIB rather than RFC1213-MIB.
(This is not a bug, but... RFC1213 should be retired, IMHO.)
                 Ken

++++++++++++++++++++++++++++++++++++++++++++++
Ken Chapman           Unisphere Networks, Inc.
Tel: +1-978-614-5322  5 Carlisle Drive
Fax: +1-978-692-9992  Westford, MA 01886
Email: KChapman@UnisphereNetworks.com
++++++++++++++++++++++++++++++++++++++++++++++



> -----Original Message-----
> From: Spencer.Giacalone@predictive.com
> [mailto:Spencer.Giacalone@predictive.com]
> Sent: Thursday, October 19, 2000 1:13 PM
> To: Chapman, Ken
> Cc: djoyal@quantumbridge.com; fred@cisco.com; mibs@ops.ietf.org;
> ospf@discuss.microsoft.com; rcoltun@redback.com
> Subject: Re: draft-ietf-ospf-mib-update-04
> 
> 
> Ken, 
> 
> Thanks. The first might be a fat finger on my part, but SMIC 
> didnt pick 
> the rest up. Let me take a look and get back to you. Imports 
> might be a 
> legacy thing. 
> 
> Spence
> 
> 
> 
> 
> 
> 
> "Chapman, Ken" <KChapman@unispherenetworks.com>
> 10/19/00 11:41 AM
> 
> 
>         To:     "'ospf@discuss.microsoft.com'" 
> <ospf@discuss.microsoft.com>
>         cc:     "'spencer.giacalone@predictive.com'" 
> <spencer.giacalone@predictive.com>, 
> "'djoyal@quantumbridge.com'" <djoyal@quantumbridge.com>, 'Rob Coltun' 
> <rcoltun@redback.com>, "'fred@cisco.com'" <fred@cisco.com>, 
> "'mibs@ops.ietf.org'" <mibs@ops.ietf.org>
>         Subject:        draft-ietf-ospf-mib-update-04
> 
> 
> Hi,
> I'm not a regular on this list and don't know a whole lot 
> about OSPF; I'm
> just an SNMP geek.
> But I found some bugs in the MIBs, and I didn't see any 
> issues raised in 
> the
> archives.
> 
> There is a bug in OSPF-TRAP-MIB: 
> The last object imported from OSPF-MIB is 
> "ospfAreaExtNssaTranslatorStatus";
> it should be "ospfAreaNssaTranslatorState".
> 
> There are some minor "conformace" issues with the MIBs as presented:
> 
> The "00" year part of the dates in the LAST_UPDATED and 
> REVISION clauses
> need to be "2000"
> (ref: RFC2578 clause 2 (top of page 5).
> 
> ospfAuthType is not included in any OBJECT-GROUP definition. 
> 
> None of the notifications in OSPF-TRAP-MIB are included in a
> NOTIFICATION-GROUP. 
> (note: don't forget to import NOTIFICATION-GROUP from SMIv2-CONF.)
> 
> While the last two are not violations per se (as far as I can 
> tell from
> reading RFC2580),
> it makes it difficult to include these items in an AGENT-CAPABILITIES
> definition.
> 
> I hope this is helpful.
> Cheers.
>                  Ken
> 
> 
> 
>