Nagios Support Forum

Our Nagios XI server has two separate network interfaces, both on a different subnet. We are using SNMP for our monitoring at the moment. Is it possible to specify multiple interfaces that Nagios would use to obtain SNMP information from other servers? We need information from servers on both subnets and we do not want to change routing within the network.

Thank you,
Kyle

The snmptrap service listens to all interfaces by default so incoming traps shouldn't be a problem. And as long as the underlying OS of the XI system is able to communicate(ping to check connectivity for example) with the monitored systems then XI shouldn't have an issue doing snmpget or snmpwalks to query those systems.

cdienger wrote:The snmptrap service listens to all interfaces by default so incoming traps shouldn't be a problem. And as long as the underlying OS of the XI system is able to communicate(ping to check connectivity for example) with the monitored systems then XI shouldn't have an issue doing snmpget or snmpwalks to query those systems.

I am referring to outbound SNMP from the Nagios server. For example. Say we want to check cpu usage on a device in vlan 60, which the nagios server can reach and vice versa, but the nagios server is configured in the web to be accessed through vlan 68. When I try to run a configuration wizard on a device in vlan 60 it does not work. This is repeatable on any of the devices in vlan 60, even when they are set up properly on their end to receive snmpv2 messages. We are running RHEL 7.5 and I noticed that when I change the primary network interface to vlan 60 it will work then, but vlan 68 machines that have previously worked using snmp will stop working.

Thanks for the clarification. I don't see another option in this case than to add a route.