icinga2/doc/14-addons-plugins.md

12 KiB

Icinga 2 Addons and Plugins

Graphing

PNP

PNP is a graphing addon.

PNP is an addon which adds a graphical representation of the performance data collected by the monitoring plugins. The data is stored as rrd (round robin database) files.

Use your distribution's package manager to install the pnp4nagios package.

If you're planning to use it configure it to use the bulk mode with npcd and npcdmod in combination with Icinga 2's PerfdataWriter. NPCD collects the performance data files which Icinga 2 generates.

Enable performance data writer in icinga 2

# icinga2 feature enable perfdata

Configure npcd to use the performance data created by Icinga 2:

vim /etc/pnp4nagios/npcd.cfg

Set perfdata_spool_dir = /var/spool/icinga2/perfdata and restart the npcd daemon.

There's also an Icinga Web 2 module for direct PNP graph integration available at https://exchange.icinga.org/icinga/PNP4Nagios

More information on action_url as attribute and graph template names.

Graphite

Graphite is a time-series database storing collected metrics and making them available through restful apis and web interfaces.

Graphite consists of 3 software components:

  • carbon - a Twisted daemon that listens for time-series data
  • whisper - a simple database library for storing time-series data (similar in design to RRD)
  • graphite webapp - A Django webapp that renders graphs on-demand using Cairo

Use the GraphiteWriter feature for sending real-time metrics from Icinga 2 to Graphite.

# icinga2 feature enable graphite

There are Graphite addons available for collecting the performance data files too (e.g. Graphios).

inGraph

inGraph requires the ingraph-collector addon to be configured to point at the perfdata files. Icinga 2's PerfdataWriter will write to the performance data spool directory.

Visualization

Icinga Reporting

By enabling the DB IDO feature you can use the Icinga Reporting package.

NagVis

By using either Livestatus or DB IDO as a backend you can create your own network maps based on your monitoring configuration and status data using NagVis.

Thruk

Thruk is an alternative web interface which can be used with Icinga 2.

Log Monitoring

Using Logstash or Graylog in your infrastructure and correlate events with your monitoring is even simpler these days.

  • Use the GelfWriter feature to write Icinga 2's check and notification events to Graylog or Logstash.
  • Configure the logstash nagios output to send passive traps to Icinga 2 using the external command pipe.
  • Execute a plugin to check Graylog alert streams.

More details can be found in this blog post.

Notification Scripts and Interfaces

There's a variety of resources available, be it different notification scripts such as:

  • E-Mail (examples provided)
  • SMS
  • Pager (XMPP, etc.)
  • Twitter
  • IRC
  • Ticket systems
  • etc.

Additionally external services can be integrated with Icinga 2:

More information can be found on the Icinga Website and the Icinga Wiki.

Configuration Management Tools

If you require your favourite configuration tool to export Icinga 2 configuration, please get in touch with their developers. The Icinga project does not provide a configuration web interface yet. Follow the Icinga Blog for updates on this topic.

If you're looking for puppet manifests, chef cookbooks, ansible recipes, etc - we're happy to integrate them upstream, so please get in touch at https://support.icinga.org.

These tools are currently in development and require feedback and tests:

Plugins

For some services you may need additional 'check plugins' which are not provided by the official Monitoring Plugins project.

All existing Nagios or Icinga 1.x plugins work with Icinga 2. Here's a list of popular community sites which host check plugins:

The recommended way of setting up these plugins is to copy them to a common directory and create a new global constant, e.g. CustomPluginDir in your constants.conf configuration file:

# cp check_snmp_int.pl /opt/monitoring/plugins
# chmod +x /opt/plugins/check_snmp_int.pl

# cat /etc/icinga2/constants.conf
/**
 * This file defines global constants which can be used in
 * the other configuration files. At a minimum the
 * PluginDir constant should be defined.
 */

const PluginDir = "/usr/lib/nagios/plugins"
const CustomPluginDir = "/opt/monitoring/plugins"

Prior to using the check plugin with Icinga 2 you should ensure that it is working properly by trying to run it on the console using whichever user Icinga 2 is running as:

# su - icinga -s /bin/bash
$ /opt/monitoring/plugins/check_snmp_int.pl --help

Additional libraries may be required for some plugins. Please consult the plugin documentation and/or plugin provided README for installation instructions. Sometimes plugins contain hard-coded paths to other components. Instead of changing the plugin it might be easier to create logical links which is (more) update-safe.

Each plugin requires a CheckCommand object in your configuration which can be used in the Service or Host object definition.

There are the following conventions to follow when adding a new command object definition:

  • Always import the plugin-check-command template
  • Use command-arguments whenever possible. The command attribute must be an array in [ ... ] then for shell escaping.
  • Define a unique prefix for the command's specific command arguments. That way you can safely set them on host/service level and you'll always know which command they control.
  • Use command argument default values, e.g. for thresholds
  • Use advanced conditions like set_if definitions.

Example for a custom my-snmp-int check command:

object CheckCommand "my-snmp-int" {
  import "plugin-check-command"

  command = [ CustomPluginDir + "/check_snmp_int.pl" ]

  arguments = {
    "-H" = "$snmp_address$"
    "-C" = "$snmp_community$"
    "-p" = "$snmp_port$"
    "-2" = {
      set_if = "$snmp_v2$"
    }
    "-n" = "$snmp_interface$"
    "-f" = {
      set_if = "$snmp_perf$"
    }
    "-w" = "$snmp_warn$"
    "-c" = "$snmp_crit$"
  }

  vars.snmp_v2 = true
  vars.snmp_perf = true
  vars.snmp_warn = "300,400"
  vars.snmp_crit = "0,600"
}

Icinga 2 has built-in check command definitions for the Manubulon Plugin Checks.

For further information on your monitoring configuration read the Monitoring Basics chapter.

You can find additional plugins at the Icinga Exchange

More details on the plugins can also be found on the Icinga Wiki at https://wiki.icinga.org

Tip

Create the best CheckCommand definition there is and send it upstream. More information can be found in Contribute Icinga 2 ITL Plugin Check Command Definitions on the Icinga Wiki. Thank you in advance!

Plugin API

Currently Icinga 2 supports the native plugin API specification from the Monitoring Plugins project.

The Monitoring Plugin API is defined in the Monitoring Plugins Development Guidelines.

There are no output length restrictions using Icinga 2. This is different to the Icinga 1.x plugin api definition.

More Addon Integration Hints

PNP Action Url

They work in a similar fashion for Icinga 2 and are used for 1.x web interfaces (Icinga Web 2 doesn't require the action url attribute in its own module).

template Service "pnp-hst" {
  action_url = "/pnp4nagios/graph?host=$HOSTNAME$"
}

template Service "pnp-svc" {
  action_url = "/pnp4nagios/graph?host=$HOSTNAME$&srv=$SERVICEDESC$"
}

PNP Custom Templates with Icinga 2

PNP automatically determines the graph template from the check command name (or the argument's name). This behavior changed in Icinga 2 compared to Icinga 1.x. Though there are certain possibilities to fix this:

  • Create a symlink for example from the templates.dist/check_ping.php template to the actual check name in Icinga 2 (templates/ping4.php)
  • Pass the check command name inside the format template configuration

The latter becomes difficult with agent based checks like NRPE or SSH where the first command argument acts as graph template identifier. There is the possibility to define the pnp template name as custom attribute and use that inside the formatting templates as SERVICECHECKCOMMAND for instance.

Example for services:

# vim /etc/icinga2/features-enabled/perfdata.conf

service_format_template = "DATATYPE::SERVICEPERFDATA\tTIMET::$icinga.timet$\tHOSTNAME::$host.name$\tSERVICEDESC::$service.name$\tSERVICEPERFDATA::$service.perfdata$\tSERVICECHECKCOMMAND::$service.check_command$$pnp_check_arg1$\tHOSTSTATE::$host.state$\tHOSTSTATETYPE::$host.state_type$\tSERVICESTATE::$service.state$\tSERVICESTATETYPE::$service.state_type$"

# vim /etc/icinga2/conf.d/services.conf

template Service "pnp-svc" {
  action_url = "/pnp4nagios/graph?host=$HOSTNAME$&srv=$SERVICEDESC$"
  vars.pnp_check_arg1 = ""
}

apply Service "nrpe-check" {
  import "pnp-svc"
  check_command = nrpe
  vars.nrpe_command = "check_disk"

  vars.pnp_check_arg1 = "!$nrpe_command$"
}

If there are warnings about unresolved macros make sure to specify a default value for vars.pnp_check_arg1 inside the

In PNP, the custom template for nrpe is then defined in /etc/pnp4nagios/custom/nrpe.cfg and the additional command arg string will be seen in the xml too for other templates.