I am trying to send a trap message from my agent to net-snmp but in my agent it looks like it's fine, however net-snmp doesn't show the message.
I want to know if net-snmp knows how to handle a trap message?
Normally, you would use net-snmp as an Agent.
Agents generally send traps, and don't receive them. They need to be able to handle get-requests, set-requests and get-next-requests, but not traps.
The system which receives SNMP traps is known as a Manager. Net-snmp is not an SNMP manager, but it contains all the (C) programming libraries needed for you to create your own SNMP manager, should you want to. There is also snmptrapd
, which can log incoming traps into a file, or pass them to a perl script for processing.
Perhaps you're indeed asking about snmptrapd
?
If traps are not logged on the receiver, normal network troubleshooting procedures apply. Snoop the traffic using Wireshark, first on the receiving system, then if needed, on the sending side. This can verify whether the traffic is dropped by the network (firewalls etc).
If you want to study the basics of the protocol, this RFC is a good starting point: "Protocol Operations for Version 2 of the Simple Network Management Protocol" http://www.ietf.org/rfc/rfc1905.txt