Skip to main content

Ganglia metric check plugin for Nagios

Project description

check_ganglia_metric is a Nagios plugin that allows you to trigger alerts on any Ganglia metric.

Installation

# pip install check_ganglia_metric

…or:

# easy_install check_ganglia_metric

Ganglia Configuration

Unless your Nagios server and Ganglia Meta Daemon are running on the same host, You probably need to edit your gmetad.conf to allow remote connections from your Nagios server.

To allow connections from nagios-server.example.com:

trusted_hosts nagios-server.example.com

To allow connections from all hosts (probably a security risk):

all_trusted on

Testing on the Command Line

First, let’s see if check_ganglia_metric can communicate with the Ganglia Meta Daemon:

$ check_ganglia_metric.py --gmetad_host=gmetad-server.example.com \
  --metric_host=host.example.com --metric_name=cpu_idle
Status Ok, CPU Idle = 99.3 %|cpu_idle=99.3%;;;;

The “Status Ok” message indicates that check_ganglia_metric is working. If you’re having trouble getting this to work, try again with verbose logging enabled (--verbose) in order to gain better insight into what’s going wrong.

Now let’s try setting an alert threshold:

$ check_ganglia_metric.py --gmetad_host=gmetad-server.example.com \
  --metric_host=host.example.com --metric_name=cpu_idle --critical=99
Status Critical, CPU Idle = 99.6 %|cpu_idle=99.6%;;99;;

We told check_ganglia_metric to return a “Critical” status if the Idle CPU was greater than 99. The “Status Critical” message indicates that it worked. Note that check_ganglia_metric uses NagAconda to parse ranges and thresholds according to the official Nagios plugin development guidelines.

To see a complete list of command line options with brief explanations, run check_ganglia_metric with the --help option.

Nagios Configuration

First, create a command definition:

define command {
  command_name  check_ganglia_metric
  command_line  /usr/bin/check_ganglia_metric.py --gmetad_host=gmetad-server.example.com --metric_host=$HOSTADDRESS$ --metric_name=$ARG1$ --warning=$ARG2$ --critical=$ARG3$
}

Now you can use the above command in your service definitions:

define service {
  service_description  CPU idle - Ganglia
  use                  some_template
  check_command        check_ganglia_metric!cpu_idle!0:20!0:0
  host_name            host.example.com
}

This will work fine until something goes wrong with check_ganglia_metric (e.g. the cache file can’t be read/written to, the Ganglia Meta Daemon can’t be reached, etc.). At that point, every service that relies on check_ganglia_metric will fail, possibly inundating you with alerts. We can prevent this through the use of service dependencies.

The first thing we need is a command definition for checking the age of a file:

define command {
  command_name  check_file_age
  command_line  /usr/lib/nagios/plugins/check_file_age -f $ARG1$ -w $ARG2$ -c $ARG3$
}

Next, we define a service which checks the age of check_ganglia_metric’s cache file. Note that in order to be truly effective, this service needs to be checked at least as (preferably more) frequently than all the other checks that rely on check_ganglia_metric:

define service {
  service_description  Cache for check_ganglia_metric
  use                  some_template
  check_command        check_file_age!/var/lib/nagios/.check_ganglia_metric.cache!60!120
  host_name            localhost
  check_interval       1
  max_check_attempts   1
}

And finally, we set up the actual service dependency. Note that I’ve enabled use_regexp_matching in Nagios, which allows me to use regular expressions in my directives. By sticking “- Ganglia” at the end of every service that relies on check_ganglia_metric, I can save myself a lot of effort:

define servicedependency {
  host_name                      localhost
  service_description            Cache for check_ganglia_metric
  dependent_host_name            .*
  dependent_service_description  .* \- Ganglia$
  execution_failure_criteria     c,p
}

Now if something goes wrong with check_ganglia_metric, only one alert will be sent out about the cache file, and all dependent service checks will be paused until you fix the problem that caused check_ganglia_metric to fail. Once the problem is fixed, you’ll need to update the timestamp on the cache file in order to put the “Cache for check_ganglia_metric” service back into an OK state (which will allow dependent service checks to continue):

$ touch /var/lib/nagios/.check_ganglia_metric.cache

Tips and Tricks

It’s possible to get a complete list of available hosts and metrics by enabling “more verbose” logging (-vv). Since the metric_host and metric_name options are required, you have a little bit of a “chicken and egg” problem here, but that’s OK. Just supply some dummy data. The plugin will error out at the end with a “host/metric not found” error, but not before it dumps its cache:

$ check_ganglia_metric.py --gmetad_host=gmetad-server.example.com \
  --metric_host=dummy --metric_name=dummy -vv

Known Issues

  • Doesn’t work with Python 2.4

Change Log

2012.02.28

  • Add –metrics_max_age option

  • Merge CHANGELOG.rst into README.rst

2011.09.09

  • Fix crash when metric name has no title

  • Add CHANGELOG.rst

2011.05.10a

  • Initial public release

License

Copyright (C) 2011 Michael Paul Thomas Conigliaro

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the “Software”), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Credits

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

check_ganglia_metric-2012.02.28.tar.gz (12.8 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page