Groups | Search | Server Info | Login | Register
Groups > comp.protocols.snmp > #217
| From | "Mr. Uh Clem" <uhclem@DutchElmSt.invalid> |
|---|---|
| Newsgroups | comp.protocols.snmp |
| Subject | GetBulkRequest partial results? |
| Date | 2015-02-04 11:49 -0500 |
| Organization | Brotherhood of Zips and OtherS |
| Message-ID | <cjf0teF2qvsU1@mid.individual.net> (permalink) |
http://tools.ietf.org/search/rfc3416#page-14
I've not used GetBulk before and am wondering about the case of partial
results. In reading the fine RFC, it seems that if the agent runs out of
space for results, it simply stops when the response for the next object
will not fit. This is simply object by object, not per iteration of the
repeated objects(conceptual row)? For example, if I'm polling five objects
(A-E) with a single non-repeater (N) with repeater count of 10 and the agent
runs out of space in the middle of the third repetition, I might expect
to get instead of
NA1B1C1D1E1 NA1B1C1D1E1
A2B2C2D2E2 A2B2C2D2E2
A3B3C3
, the later cleanly breaking at a repetition (row).
What's the way to continue to traverse the table? Seems like doing another
GetBulk using the instances from the last complete repetition should work.
possibility. It feels like it should be possible to continue by doing
something like a GetBulk with non-repeating D2E2 and repeating A3B3C3D?E?,
but there's really no way to come up with the instances for D and E, is
there? Am I missing something?
--
Clem
"If you push something hard enough, it will fall over."
- Fudd's first law of opposition
Back to comp.protocols.snmp | Previous | Next | Find similar
GetBulkRequest partial results? "Mr. Uh Clem" <uhclem@DutchElmSt.invalid> - 2015-02-04 11:49 -0500
csiph-web