I'm going to extend the Net-SNMP agent version 5.2.1.2 running on an embedded hardware (ARM9 processor) with a private enterprise MIB subtree. I read the documentations (tutorials, man pages and so on) on the web site, but I can't dedice between pass, dlmod and perhaps AgentX.
I'll write the subagent from scratch because I don't have now any code that manages the objects I need to export by SNMP. I tried with pass option in snmpd.conf, writing a small program that reads the command line parameters (-g, -n or -s) and prints to standard output the result of the operation. It's very simple and I built a small example in 5 minutes... no more. But I know there are other methods to extend the Net-SNMP agents, for example dynamically loadable modules and AgentX subagent. In both cases I think I need mib2c to generate a template C code to fill with details about my objects. Surely these two methods are more complex to understand and write. So the question is: what are the main advantages to use dlmod or AgentX approach against the external stand-alone application that reads from command line parameters and writes to standard output? I'm quite sure dlmod and AgentX methods are better than pass, but I didn't find the reason. I'm aware starting coding following a way and realizing only in the future that the way is wrong... Thank you for your answers. ------------------------------------------------------------------------- SF.Net email is sponsored by: The Future of Linux Business White Paper from Novell. From the desktop to the data center, Linux is going mainstream. Let it simplify your IT future. http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4 _______________________________________________ Net-snmp-coders mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
