> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On
> Behalf Of Vincent Royo
> Sent: Wednesday, April 02, 2008 4:45 AM
> I want to use an agentx on some nodes (pc) and make it communicate
> over tcp with a master agent on one other.
>
> I have the same data on each node, like number of disks, capacity, the
> status of a dynamic set of services, etc.
You probably shouldn't use AgentX, then. The general idea behind
AgentX is to allow an agent to be extended. E.g., you add an ATM interface
card for which the vendor provides an AgentX MIB implementation - this MIB can
be connected to any master agent which supports AgentX. This works really well
when multiple MIBs are added - if those MIBs only worked with certain master
agents, you would only be able to manage certain combinations of equipment.
AgentX was not intended for distributed MIBs. It can be done, though,
if you really understand the protocol. I do not know what support for such is
available in net-snmp.
> The same mib is installed on all nodes. I think the correct way is to
> use tables.
I think the correct way is to put a master agent on each node and
connect each MIB to its own master.
> I create the table in the mib and create the generic code with mib2c;
> what is the structure of the code generated by mib2c when I use
> mib2c.mfd.conf?
I don't know mib2c, myself. I should have learned by now, but... old
dog, you know.
HTH,
Mike
-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
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
