tanto 1.1

Takes monitoring data from Nagios-plugins to push with NSCA (Nagios or Icinga) or WS-Shinken.

• Github repository

• PyPI page

Overview

To monitor the servers with Shinken, Nagios or Icinga, system administrators usually configure active checks of the monitored servers. It means the monitoring system must have a direct network access to the monitored server.

With Tanto, this model is reverted, it does passive checks of your monitored servers. You don’t need to open an access from your monitoring system to the monitored servers.

The main use case of this tool is to monitor distant servers in a complex network where you can’t connect directly for security reasons, typically with a large client. You need only to open NSCA or HTTP(S) port from the monitored server to the monitoring system.

Moreover, if you don’t have a monitoring server, Tanto has a standalone mode. You will receive alerts via e-mail directly.

Technical details

The behaviour of this tool is simple:

1. The program is started by cron.

2. It retrieves monitoring data (1) from nagios-plugins.

3. Finally, it pushes the data (2) via NSCA (Shinken, Nagios, Icinga), HTTP(S) (Shinken) to the monitoring server or via SMTP.

4. (WS-Shinken only) If the connexion between the monitored server and the monitoring system is broken, data is kept in a cache file, to be re-send the next run.

Warning: WS-Shinken support isn’t finished.

How to install ?

You install as you wish, via sources, pip or Debian package. If you use another distribution, you can use the Python package or contribute to publish a package for your distribution.

Debian

We provide a Debian package source in the debian/ folder.

Python package

pip install tanto

Usage

• Default config files location: /etc/tanto/

• Default cron file location: /etc/cron.d/

1. Configure the nagios plugins you want to use in inputs/nagios_plugins.cfg:

   1. Define the nagios plugins path with the path option in [default_settings] section.

   2. The name of each section is the nagios plugin command.

   3. Each setting is a CLI option of the nagios plugin.

2. Adapt the checks frequency in: /etc/cron.d/tanto.

Standalone mode

1. Configure e-mail notifications in outputs/email.cfg.

With a monitoring server

1. Configure passive checks on your monitoring system:

   1. Shinken

   2. (PDF) Nagios

   3. Icinga

2. Fill in the credidentials for your monitoring system in outputs/nsca.cfg or outputs/ws_shinken.cfg.

CLI options

Launch: tanto –help You can override the location of each config file with a CLI parameter.

Debug

By default, it uses cron syslog and console for the logs. You can customize this behaviour in logging.ini: http://docs.python.org/library/logging.config.html#configuration-file-format

Support

Community support is provided via Github: https://github.com/Eyepea/tanto/issues

You must provide the error log with your issue.

If you need professional support, please contact Eyepea: http://www.eyepea.eu/

For general questions or contributions, you can contact me via my Github acccount: https://github.com/GMLudo

Special thanks

The Shinken community in general, and Jean Gabès in particular for his help and support.

Release History

1.1 (2014-05-12)

• Add e-mail notifications for simple infrastructures without a monitoring server.

• Move config files in input/ and output/ to be easier to understand.

• Reduce command line arguments for config files.

1.0 (2014-05-06)

• Change nagios_plugins.cfg syntax to give directly the command line to launch.

• Add servers list parameter in nagios_plugins.cfg to define the servers list that must receive configuration.

• Logging improvements

• Rename eyepea_monitoring_agent to Tanto, a short blade in the same family as Shinken.

0.9 (2012-08-14)

First public release.