[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: What is "SET ROW"



I believe that the idea for this feature came from
Andy Bierman's complaint that one of the reasons
snmp SET's aren't more widely implemented is the need
for the agent to accept pieces of a row in any order.

The idea is to introduce a new row-status-like TC which
lacks create-and-wait and to push its use in all new mibs
(and perhaps in any updates of old mibs).  This combined with
either oid suppression or some other technique which allows
a whole row to fit into a single pdu would allow these new mibs
to be implemented much more easily without the need to
maintain state between pdus.

So, say a new revision of the rmon mib were released which
used the "SETROW" TC or whatever it is going to be called.
Then an implementer could assume that any set operation that
created a row would have everything that was need to create
that row (and, I guess, could default any omitted columns).

The problem is solves is difficulty of implementation of SET's.

The only problem I could see it creating is people might mis-apply it
and implement old mibs which used earlier an version of the row-status
TC (with create-and-wait) as if they could assume that all varbinds
for a row would always be in a single PDU.

At 12:00 PM 3/27/2001, you wrote:
>HI,
>
>Would someone PLEASE explain:
>  1) what a SETROW operation would do,
>  2) what problems does it solve
>  3) what problems does it create
>
>Thanks,
>/david t. perkins