Ascend Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

(ASCEND) SNMP MIBs, Different Equipment, Different Revisions

The following is offered to the group for comments prior to a 
formal request to Ascend.  I have been "management" long enough
to know that my grasp of the details is (at best) shaky in places, 
so I need a consensus.

Most companies create a private MIB tree that looks like this:

iso(1).org(3).dod(6).internet(1).private(4).enterprises(1)....

        companyX(xxx).
                productA(aaa)...
                productB(bbb)...
                productC(ccc)...

...and within each product, one can have a complete set of mib files
unique to that product (in some cases, one must be willing to do some 
editing and renaming of the mib files if the vendor does not publish 
a complete set for each product).

Ascend has a single set of MIB files that are supposed to support all 
Ascend products (and Ascend does not have a "product" level within 
their tree).  This appears to be a problem in cases like the following:

        1) Given the set of MIBs published in September,
           take two Pipeline 130-FT1s.  Configure them
           in an "identical" manner (except for the
           DCLIs and IP Addresses required to be unique
           for each Pipeline).  Pay special attention to
           the Pipeline "20-700" SNMP trap settings, and
           make sure that both Pipelines "know" the same
           SNMP manager address.

        2) Now upgrade one Pipeline to 5.1A, and leave the
           other Pipeline at 5.0Ap1.

        3) Now use any SNMP software to simply try to pull up the 
           interface table for each pipeline:

                3a)  The Pipeline at 5.1A gives us entries for:

                        Console 1
                        BRI Slot 1 Line 1                
                        Nailed T1
                        ie0
                        wan0 - wan10
                        wanidle0
                        tunnel0
                        mcast0
                        rj0
                        bh0
                        local
                        lo0

                3a)  The Pipeline at 5.0Ap1 gives us entries for:

                        Console 1
                        BRI Slot 1 Line 1                                       

                     ...and nothing else.  Not even an entry
                     for the fractional T-1 Wan port.  Not even
                     an entry for the ethernet port.

        If one reads the mib file, one expects both devices to
        give similar information in this very basic table.

        If this is the situation with the ascend-supplied RFC mibs, 
        and a simple revision difference, what happens when we have 
        a mix of Ascend products AND an mix of revisions, all expecting 
        us to have different mib revisions?

        What happens when we have two vendors, each who make use of a
        different "RFC mib" that they supply in subset form?

        It would appear that Ascend needs to add the following levels 
        to their mib tree:

             ...ascend(529).
                  product(1-x).
                    revision_major_num(4 and up).
                      revision_suffix(0-9???).
                        HERE STARTS THE TABLES

       If this were done, one could manage a random group of different
       Ascend products at different revisions, and be able to work with
       older mibs that are appropriate for older versions. 

        Where I come from, we call this "working", and we call the
        example above "broken".  

        Comments?

 Surfing the electromagnetic spectrum, looking for that perfect Hertzian wave

james fischer                                jfischer@supercollider.com

++ Ascend Users Mailing List ++
To unsubscribe:	send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd:	<http://www.nealis.net/ascend/faq>