Nagios Support Forum

My organization is evaluating NagiosXi as a monitoring solution and I have been tasked with looking at it to see if it can meet our needs to monitor the state of connections to different peers configured on a monitored host. The monitored host sends SNMP trap to reflect the connection up/down status.

I found the SNMP Trap wizard under the Configuration wizards in NagiosXI which I thought would meet our requirements. I have installed it by following the steps provided in the following doc:https://assets.nagios.com/downloads/nag ... ios_XI.pdf.

I was able to configure the NagiosXI host to receive the traps from the monitored node and to send notifications with the trap information in it. However, I was not able to get the overall SNMP Trap service to reflect a status in the web-gui the way I want.

Let me describe one of our use cases.

The monitored node establishes connections to 10 peers of one type labeled as PeerA-1 through PeerA-10, and connections to 250 peers of a second type labeled as PeerB-1 through PeerB-250. I need to monitor these individual connections on the node. The node raises and clears alarms to reflect the connection status and generates corresponding SNMP traps. /etc/snmp/snmptt.conf has been configured as follows for this:

Connection to PeerA-1 dropped : EVENT peerConAFault .1.3.6.1.4.1.17402.110.2.1.1.2.0.3 "Status Events" Critical
Connection to PeerA-1 restored: EVENT peerConAFault .1.3.6.1.4.1.17402.110.2.1.1.2.0.103 "Status Events" Normal

Connection to PeerB-1 dropped : EVENT peerConBFault .1.3.6.1.4.1.17402.110.2.1.1.2.0.4 "Status Events" Critical
Connection to PeerB-1 restored: EVENT peerConAFault .1.3.6.1.4.1.17402.110.2.1.1.2.0.104 "Status Events" Normall

The trap includes an OID which designates the type of event. It also includes connections details including the peer name, the ipAddress and the port number.

I want NagiosXi to reflect the overall status "OK" only when all the connections on the monitored node are up.

I also want the ability to quickly drill down to a list dropped connections based on the peer type so corrective action could be taken quickly on the monitored node. I want to avoid configuring individual connections in NagiosXI as that will be a considerable task for the current 260 connections which may increase significantly in the future.

Thanks for your help!

Greetings!

You may want to look into the BPI feature under Home > Details > BPI. Here you could create a BPI group of all the SNMP traps and designate all as essential members and give the group high priority. Once this is configured you'd see the group under the High Priority tab which would turn critical if one device had a problem and would allow you to quickly see which device it was. https://assets.nagios.com/downloads/nag ... BPI_v2.pdf covers more on bpi including how to setup a bpi check to alert you if the group is having a problem.