Nagios Performance Graphs not showing up

[lovespie]

CPU Usage, Load Average, Memory Usage, Swap Usage, Total Processes, and Zombie Processes are not showing up in the Nagios Performance Graphs. In its place is just "No results found. Inspect ..."

Warning and Critical Alerts, and Top 10 Service Notifications are working so I don't think the problem is MK Livestatus.

I think it has something to do with what I'm naming the service description but I don't know what the name should be or which file or tag to edit. I've tried renaming the service description to "Memory Usage", "Memory usage", "mem_used", and "mem_total", but every time the result is the same.

For now the plugin used is check_mem.pl

From service-perfdata:
1377833717 src_host="localhost" perfdata="SERVICEPERFDATA" name="Memory Usage" severity="CRITICAL" attempt="3" statetype="HARD" executiontime="0.032" latency="0.186" reason="CRITICAL - 96.7% (588532 kB) used!" result="TOTAL=608480KB;;;; USED=588532KB;;;; FREE=19948KB;;;; CACHES=419540KB;;;;"

From props.conf
EXTRACT-mem_used = Memory usage=(?\d+.\d+)Mb;
EXTRACT-mem_total = Memory usage=\d+.\d+Mb;\d+.\d+;\d+.\d+;\d+.\d+;(?[^;]*)(?=)

[cvervais]

I'm running into the same thing and I'm wondering if it's because of the src_host= field. On my install it's localhost as well. So, looking at the graphs it seems to be expecting the name of the host there that you're trying to view data for.

For example, for CPU usage for one of my hosts it's using this search query:

index=nagios (sourcetype="nagioshostperf" OR sourcetype="nagiosserviceperf") src_host="wp-admin-prod" | bin span=5m _time | table *

But, since there are no events in my nagios index where src_host="wp-admin-prod", no results are going to be found. In all my events the value for src_host is localhost.

Here are the command definitions I used in my nagios commands.cfg file:

'nagios-process-host-perfdata' command definition

define command{
command_name nagios-process-host-perfdata
command_line /usr/bin/printf "%b" "$TIMET$ src_host=\"$HOSTNAME$\" perfdata=\"HOSTPERFDATA\" hoststate=\"$HOSTSTATE$\" attempt=\"$HOSTATTEMPT$\" statetype=\"$HOSTSTATETYPE$\" executiontime=\"$HOSTEXECUTIONTIME$\" reason=\"$HOSTOUTPUT$\" result=\"$HOSTPERFDATA$\"\n" >> /usr/local/nagios/var/host-perfdata
}

'nagios-process-service-perfdata' command definition

define command{
command_name nagios-process-service-perfdata
command_line /usr/bin/printf "%b" "$TIMET$ src_host=\"$HOSTNAME$\" perfdata=\"SERVICEPERFDATA\" name=\"$SERVICEDESC$\" severity=\"$SERVICESTATE$\" attempt=\"$SERVICEATTEMPT$\" statetype=\"$SERVICESTATETYPE$\" executiontime=\"$SERVICEEXECUTIONTIME$\" latency=\"$SERVICELATENCY$\" reason=\"$SERVICEOUTPUT$\" result=\"$SERVICEPERFDATA$\"\n" >> /usr/local/nagios/var/service-perfdata
}

So, it's likely got to do with $HOSTNAME$ macro not being set properly by nagios but, I'm at a loss to get any further.

[cvervais]

Figured it out, it was a rookie mistake on my part. In the nagios.cfg I had host_perfdata_command and service_perfdata_command set incorrectly. The values were process-host-perfdata and process-service-perfdata which were defaults. The command definitions for those in the commands.cfg file don't print out the right hostname and aren't the right commands to be running anyways.

Per the documentation (Yeah I didn't RTFM carefully) the commands the instructions specify are nagios-process-host-perfdata and nagios-process-service-perfdata respectively. Once I updated nagios.cfg to these all is well

[lukeh]

Hi 🙂

I have removed the name field as a requirement from the dashboard in Splunk for Nagios version 3.0.0, please upgrade and let me know how you go 🙂

All the best,

Luke 🙂