On Fri, Jul 16, 2010 at 8:21 AM, Dave Shield <[email protected]>wrote:
> On 16 July 2010 13:11, <[email protected]> wrote:
> > It times out while we are walking through mib file we created for our
> > hardware.
>
> Then the problem must lie within the code that you have
> written to implement your MIB.
>
> This is the first time that you have mentioned a private MIB.
> Up to now, you have been giving the impression that you're
> having problems with the standard MIBs.
>
... snip ...
> > Or this break is creating the problem?
>
> The various MIB files (and the distinction between standard and
> user-defined) is purely an administrative convenience for us humans.
> The agent/apps/protocol all work with a combined OID tree containing
> "everything they know about"
>
But I have encountered problems with sub-agents written by my co-workers
that did not 'do the right thing' at the end of tables and their 'assigned
block'
of the OID tree.
They did two things wrong, they:
a) didn't always return the correct 'next' OID within their block, when
asked.
b) didn't return the 'right' indication when asked for the next OID after
the last
thing _in_ their part.
As a result, when walking a device with their sub-agent included, I found
that parts of the tree (including some standard parts like sysName)
got broken (ie were now missing)!
When they fixed their code... all normality was restored.
So despite the fact that its the agents don't walk the tree, they DO have to
return
the RIGHT value for the next OID. Check your code.
Fulko
------------------------------------------------------------------------------
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
_______________________________________________
Net-snmp-users mailing list
[email protected]
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/listinfo/net-snmp-users
