On Mon, Mar 23, 2009 at 2:58 PM, Mike Ayers <[email protected]> wrote:
> > From: Fulko Hew [mailto:[email protected]]
> > Sent: Monday, March 23, 2009 11:23 AM
>
> >> On Mon, Mar 23, 2009 at 1:31 PM, Mike Ayers
> >> <[email protected]> wrote:
> >>
> >>
> >> > From: Fulko Hew [mailto:[email protected]]
> >> > Sent: Saturday, March 21, 2009 6:43 AM
> >>
> >> > if I have multiple IP addresses on my box, I'd like to (need
> >> > to) discriminate between
> >> > one request sent to 1.2.3.4 and a different request
> >> sent to 5.6.7.8
> >>
> >> The simple, effective solution is to run a
> >> different agent on each interface.
>
> > That could be hundreds, or thousands in my case!
>
> ???!!! Perhaps you should explain the use case. Why are there
> hundreds or thousands of interfaces (note: your original question implies
> two interfaces) on your system? That's a lot even for a router!
Your right. My example was only an example of what I was trying to
accomplish,
not the context in which I was trying to accomplish it.
> > It seems to me that its not the most efficient way to do things.
>
> The "thing" you are doing does not appear to be a use case for the
> master agent, so it should not be surprising if there is not an efficient
> way to do it. You may not be using the right tool for the job. Source
> network addresses are, by design, not considered relevant to processing
> management requests.
Not the source address, but the destination addresses (in my case).
What are you trying to accomplish?
I am making a 'network simulator', where a simple application (in Perl)
will cause Net-SNMP to 'forget'
all of the variables it would normally answer to, and replace them with
internally generated/supported
variables. This way I can generate 'artificial' responses to any SNMP
requests, and simulate 'any'
kind of device. The second half of the project will be to use IP aliases on
the interface so that this
one box will respond to a (large, configurable) number of addresses, where
each address will be
a different device... each with possibly a different set of supported OIDs,
and possibly a different
set of 'fake' data.
Rather than writing my own SNMP receiver (that supports all the different
protocol versions), I thought
I could simply write a Net-SNMP plugin. I can, but my remaining issue is
this IP address thing,
and I can create a simple workaround (if I have to), but I'd rather be able
to do it 'right', in a
supportable manner.
I need this to stress/regression test a number of network management
platforms we have/use.
Yes, I know there are some commercial packages out there that will do this,
but... you know...
for a variety of reasons, I am doing this myself.
------------------------------------------------------------------------------
Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are
powering Web 2.0 with engaging, cross-platform capabilities. Quickly and
easily build your RIAs with Flex Builder, the Eclipse(TM)based development
software that enables intelligent coding and step-through debugging.
Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com
_______________________________________________
Net-snmp-coders mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
