03-04-2014 06:15 AM
I have installed NNMi v9.20 on RedHat 6.5 in order to monitor SNMP traps, but it doesn't work. I have unchecked "Discard Unresolved SNMP Traps and Syslog Messages" under Configuration -> Incidents -> Incident Configuration. I have loaded the relevant MIB and added the trap under Configuration -> Incidents -> SNMP Trap Configurations.
SNMP polling seems to work as I can view the Node Component Gauges under Inventory -> Nodes.
According to the netstat command, the process "ovjboss" is listening on UDP port 162, which is the SNMP trap port.
If I run "/opt/OV/bin/nnmtrapdump.ovpl -short -t" I get nothing, but "tcdump -i eth0 udp port 162" proves the trap is indeed entering the system.
I see nothing strange in the log files under "/var/opt/OV/log/nnm" but one thing is strange, although I don't know the significance of it:
[root@darkstar nnm]# cat /var/opt/OV/log/Ovfqdn_log.txt FATAL: Permission denied at /opt/OV/nonOV/perl/a/lib/site_perl/5.8.8/OVfqdn.pm line 148. Failed to get IP address:Permission denied at /opt/OV/nonOV/perl/a/lib/site_perl/5.8.8/OVfqdn.pm line 148. [root@darkstar nnm]#
The overall status seems to be fine:
[root@darkstar nnm]# /opt/OV/bin/ovstatus object manager name: OVsPMD state: RUNNING PID: 6050 last message: - exit status: - object manager name: pmd state: RUNNING PID: 6051 last message: Initialization complete. exit status: - object manager name: nnmaction state: RUNNING PID: 6544 last message: Initialization complete. exit status: - object manager name: nmsdbmgr state: RUNNING PID: 6052 last message: Database available. exit status: - object manager name: ovjboss state: RUNNING PID: 6080 last message: Initialization complete. exit status: -
So... how can I troubleshoot this?
03-04-2014 10:06 AM
Ok, hopefully your Linux enviro's are set, that do you see when you run ovstatus -v ovjboss?
03-04-2014 10:23 AM
I didn't set any Linux environment variables myself. Do I need to? The status is this:
[root@darkstar ~]# /opt/OV/bin/ovstatus -v ovjboss object manager name: ovjboss state: RUNNING PID: 6080 last message: Initialization complete. exit status: - additional info: SERVICE STATUS CommunicationModelService Service is started CommunicationParametersStatsService Service is started CustomPoller Service is started IslandSpotterService Service is started ManagedNodeLicenseManager Service is started MonitoringSettingsService Service is started NamedPoll Service is started NmsApa Service is started NmsCustomCorrelation Service is started NmsDisco Service is started NmsEvents Service is started NmsEventsConfiguration Service is started NmsExtensionNotificationService Service is started NmsTrapReceiver Service is started NnmTrapService Service is started PerformanceSpiConsumptionManager Service is started RbaManager Service is started SpmdjbossStart Service is started StagedIcmp Service is started StagedSnmp Service is started StatePoller Service is started TrapConfigurationService Service is started TrapPropertiesService Service is started TrustManager Service is started [root@darkstar ~]#
03-05-2014 07:41 AM
Thanks! I haven't progressed anything yet. I have tried to monitor the suspicious processes with strace to see what goes wrong after a trap comes in, but I haven't seen any action at all.
I have no good ideas how to troubleshoot further :/ But if I happen to solve it, I'll post my findings here.
03-05-2014 04:04 PM
Ok so it was a stupid error afterall. I didn't realize Redhat Linux comes with a firewall enabled by default. So I added a line
-A INPUT -p udp --dport 162 -j ACCEPT
into /ets/sysconfig/iptables and ran "service iptables reload".
That was it.