Hi,

There are a few issues that need to be corrected and clarified before
submitting this document to the IESG. There may be more, but I did not
have the time to perform a full MIB Doctor review, this is what I found
at a first pass reading of the document.

- Runing smilint indicates a number of problems - most derive from the
fact that indices of many table are of the SYNTAX of Integer32 with no
range restriction
- LispAddressType TC is defined as a four-tuple with several variants -
how can it have 0 as its lower size? 
- Having a REFERENCE clause that says "[LISP]" is not useful. People may
use the MIB module but not the RFC, and even if they know what the
reference is just one high level reference is useless. To be useful to
implementers REFERENCE clauses of specific objects must point to the
exact paragraphs that define the origin of the MIB definition in the
LIPS RFC 
- Some of the Integer objects have obviously a range, but this is not
specified - for example lispMapCacheLocatorRlocPriority or
lispMapCacheLocatorRlocWeight
- The indexation of lispIidToVrfTable seems broken - if the value of
VPNIdOrZero is zero because a VPN ID could not be determined, how are to
rows in such situation distinguished? 
- There is no discontinuity indicator for the (many) counter objects in
the MIB module. See section 4.6.1.2 in RFC 4181 for some advice on this
issue
- IANA Considerations - IANA must allocate a branch under mib-2 for this
MIB module (marked xxx in the MIB module)

Regards,

Dan
 



> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf
Of
> Sander Steffann
> Sent: Wednesday, August 29, 2012 12:36 PM
> To: Terry Manderson
> Cc: LISP mailing list list
> Subject: Re: [lisp] WGLC for draft-ietf-lisp-mib-05
> 
> Hi,
> 
> > I have not seen any review nor comments made of this document in
this
> > WGLC by anyone in the WG.
> >
> > I am extending this last call by a further 14 days.
> >
> > The extended LC will end on Tuesday the 11th of September. Please
WG,
> > cast your eyes over this document, it cannot progress without
adequate
> review!
> 
> I don't see anything wrong with the document. It would have been nice
if
> notifications had been defined, for example for (certain) changes to
> lispEidRegistrationTable, lispMappingDatabaseTable and
> lispMappingDatabaseLocatorTable. But everything that is defined in the
> document looks good!
> 
> Thanks,
> Sander
> 
> _______________________________________________
> lisp mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/lisp
_______________________________________________
lisp mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lisp

Reply via email to