[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: OID compression/suppression support
On the following....
At 10:21 AM 9/7/2002 -0400, Michael Daniele wrote:
>A couple of comments/questions.
>Kristine Adamson wrote:
>> We have had a lot of complaints from network management apps about their
>> ability to get all the data they want from SNMP efficiently.
>This is a very important point, which I'd like to see more
>fully explored. Some of the eos drafts list specific problems
>with the existing protocol, but I don't think I've seen anything
>about actual functional requirements from applications.
I don't know where you have been for the last few years.
What has been happening over and over again is the attempt to
separate out the actual needs from developers that write
well written management apps versus the needs of others.
>You know, something like
> - Retrieve an entire table
> - Retrieve a particular row
> - Set a value or values in a particular row
On the above, any management app writer working for me that wanted
an operation to "retrieve a whole table" or "retrieve a row" would
be sent back to school for retraining. This request keeps coming
up over and over again, and I really wish some one would figure
out why people want this and let me know the reason.
The third item is a fundamental part of SNMP-based management, and
if it is not available, (or management app writers don't think its
available) then either a new SNMP library is needed, or the writer
needs to be sent back to school.
>I suspect some you wouldn't see are:
> - Set values in multiple rows concurrently
> - Retrieve the lexi-next instance of anything
The operation of "set values in multiple rows concurrently", while
not needing to be used very often, probably needs to be available.
And the operation of "retrieving in lexi-order", is a needed
mechanism of iteration. There is no way to write mgmt apps without
iteration. Now, if the MIB designer did a poor job of on
the indexing design, then the result can be a slow mgmt app.
/david t. perkins