Re: Integrating net-SNMP agent into my application

OoO En  ce début  d'après-midi nuageux du  samedi 03 janvier  2009, vers
14:36, "Bart Cornelissen" <[email protected]> disait :

>> 1) Subagent    http://www.net-snmp.org/wiki/index.php/TUT:Writing_a_Subagent
>> 2) Shared object
>> http://www.net-snmp.org/wiki/index.php/TUT:Writing_a_Dynamically_Loadable_Object
>> 3) Chaging the master agent
>> http://www.net-snmp.org/wiki/index.php/TUT:Writing_a_MIB_Module

> If I understand it right, these tutorials describe how I can plug my
> own MIB module, with knowledge of how to access my server
> application's variables, into the agent. Since I don't want to (and I
> guess I'm not supposed to either) incorporate my entire server into
> this module, the MIB module will have to communicate with my server
> application somehow.

Using  subagent (AgentX),  you will  write something  that will  talk to
snmpd using  AgentX protocol. I  think this is  what you want to  do and
this is usually  the way to go. The subagent  will be tightly integrated
into your  application, so it  will be easy  to access any  variable you
need or call  any function you need without  building a protocol between
this subagent and your application.

NetSNMP code will handle the communication between your subagent and the
master agent (which is snmpd).
-- 
panic("esp: what could it be... I wonder...");
	2.2.16 /usr/src/linux/drivers/scsi/esp.c

------------------------------------------------------------------------------
_______________________________________________
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