Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

RE: [802.3ae] FW: WIS MIB list of open issues and recommendationsfor resolution




On Thu, 27 Dec 2001, Tom Alexander wrote:
> Roy, it looks like Dan's explanation is in general agreement with
> the point you made in your e-mail as well. The WIS and SONET/SDH
> MIBs are being kept separate for the reasons you mentioned.

On Thu, 27 Dec 2001, Roy Bynum wrote:
> I would take a different tack on the issue of "optional" WIS MIB
> items.  I would prefer the stance that the MIB for the WIS is
> restricted to the items that are included in the current draft,
> and not include any additional "SONET" or "SDH" MIB items.  SONET
> and SDH are specific and different protocol standards from 802.3ae.
> The IETF should not confuse these standards, as they originate from
> very different standards organizations[,] by trying to get creative
> with the MIB definition for 802.3ae.

On Thu, 27 Dec 2001, Dan Romascanu wrote:
> The problem is that some of the objects that the authors of the
> SONET MIB decided that may remain optional for their management
> model are mandatory for a full definition and understanding of
> the WIS interface behavior.  This is the reason for defining a
> separate compliance statement that is stricter than the
> compliance statement for a 'regular' SONET MIB implementation.

Further elaboration may be helpful.  The draft SNMP WIS MIB in
http://www.ietf.org/internet-drafts/draft-ietf-hubmib-wis-mib-01.txt
defines some new WIS-specific objects but also incorporates by
reference a specific subset of the objects in the SONET/SDH MIB.
That subset of course exludes objects that are not relevent to the
WIS (e.g. objects for managing SONET VTs), but it also includes as
mandatory objects some that are considered conditionally mandatory
or optional in the SONET/SDH MIB compliance statement (e.g. near-
and far-end line and path layer objects).  All items listed in the
draft SNMP WIS MIB compliance statement are mandatory, including both
those that map to objects in the 802.3ae pWISBasic package and those
that map to objects in the pWISOptional package (see Section 3.6 of
the draft for details).  It was the consensus of the design team (and
of the meeting participants at Salt Lake City) that all these objects
need to be present to ensure effective management of the WIS.

Regards (and best wishes for 2002),

Mike Heard