🚀 go-pugleaf

In article <1goqabINN84h@meaddata.meaddata.com> rfrink@meaddata.com (Rick Frink) writes:
>   Why isn't the object 'collisions' part of either MIB I/II ?  It seems
>   very basic, yet essential.  ( If collisions is not relevant, then either
>   populate with something similar or nothing ).
>

Back in the Mid '80s a bunch of us at Wollongong (at the time) set about
collecting the list of variables to be part of the 1st general
purpose SNMP MIB. One of the requirements for the items related to 
the interface was that they be generic to all interfaces as much as
possible. Collisions are meaningfull on ethernet, but not on many
other media. In the years since then I've implmented many agents, and 
about 50% of the if table devices turn out to be ethernet. I guess you
could debate if one should implement a variable that only applies
about half the time, but keep in mind that we didn't know how long
ethernet would be popular as a media (we were also using token ring
and fiber). Even with hindsight, I think it was the right decision.

You should also note that MIB-II allows for extensions to the if
Table via the "ifSpecific" object. There is an ethernet MIB which maps
into that which, I assume, includes collision counts.


>Rick Frink                                              (513) 865-1645
>Mead Data Central                             Telecomm/Campus Networks
>P.O. Box 933                                       rfrink@meaddata.com
>Dayton, Ohio  45401                          ...!uunet!meaddata!rfrink
>

======================================================================
John Bartas                     | Disclaimer - I don't speak for Kalpana
Pricipal Software Engineer      | in any official capacity. I'm just
Kalpana, Inc. (408)988-1600x141 | an employee who reads the net.

"We have met the enemy, and he is us." -Pogo