You will need a Nagios Core 4.0 server or newer server installed and running to invoke nagiostats. Older versions do include the utility, but there is not quite as much information that is returned.

If you would like to run the mrtg graphing as well, which is highly recommended, you should have mrtg and its helper program indexmaker installed on your system. If you are already graphing other things with mrtg, don't worry, this recipe should not interfere with that.

The recipe does not assume any familiarity with mrtg, but if you have any problems with it, you can consult its online documentation at http://oss.oetiker.ch/mrtg/doc/index.en.html.

You should also have access to the sources from which your installation of Nagios Core was compiled. If you need to retrieve the sources again, you can download them again from Nagios Core's website, https://www.nagios.org/.

In this case, you will need to run ./configure again to generate the file required, sample-config/mrtg.cfg.

We can invoke nagiostats itself in one step whenever we want to get some statistics about the server's performance, with the following command:

With this done, if the cron task is installed correctly, we should start seeing data being plotted over the next few hours.

The statistics provided by nagiostats provides both performance data about Nagios Core itself, such as how long it takes to complete its round of checks of all objects and the average time it takes per check, as well as data such as the number of hosts in various states. By default, running it will return the data in a terse but human-readable format; you can get a good idea of the meaning of each of the fields by running it with --help, as suggested in the recipe.

The mrtg.cfg file included in the Nagios source distribution, which is tailored to your particular system at the ./configure time, contains example definitions of mrtg graphs that parse the data retrieved from nagiostats. These are not the only possible graphs using the information provided by nagiostats, but they are useful examples.

The data used is the same as the data you read if you've invoked nagiostats from the shell, but the format is slightly different. If you want to see the data being passed to mrtg by nagiostats, you can run it with the --mrtg option and nominate fields to be included in the output with --data, for example:

# /usr/local/nagios/bin/nagiostats -c /usr/local/nagios/etc/nagios.cfg --mrtg --data=AVGACTSVCPSC,AVGPSVSVCPSC,PROGRUNTIME,NAGIOSVERPID
4
0
0d 0h 5m 29s
Nagios 4.1.1 (pid=22548)

The recipe's call to indexmaker is a separate program that builds an index.html file with links to all the graphs, just for convenience. Like the mrtg call, it refers to the /usr/local/nagios/etc/mrtg.cfg configuration file to figure out what it needs to do.

Once you're happy with the way your graph's web pages are being displayed, you might like to consider including them in your Nagios Core sidebar. This can be done by editing /usr/local/nagios/share/side.php and adding a new link to the System section, perhaps named Performance Reports, below the Performance Info link. The new line might look something like this:

<li><a href="/nagios/stats/" target="<?php echo $link_target;?>">Performance Reports</a></li>

This would make a link to the graphs show up in the web interface as follows:

If you like what mrtg does with this data, you might like to look at Cacti, which is a very helpful frontend to rrdtool, which is similar to mrtg. It will allow you a lot of flexibility to define graphs, although it takes a while to learn. You can find more information about it at http://www.cacti.net/.

Another idea if you're interested in more graphing for Nagios Core performance and state data is the NagiosGraph extension, which is discussed in the Tracking host and service states with NagiosGraph recipe in Chapter 11, Automating and Extending Nagios Core.