https://github.com/jedi4ever/nagios2graphite

Forwards nagios internal metrics to graphite host (UDP)
https://github.com/jedi4ever/nagios2graphite

Last synced: about 1 month ago
JSON representation

Forwards nagios internal metrics to graphite host (UDP)

• Host: GitHub
• URL: https://github.com/jedi4ever/nagios2graphite
• Owner: jedi4ever
• Created: 2012-09-28T12:56:07.000Z (about 12 years ago)
• Default Branch: master
• Last Pushed: 2012-09-28T18:45:14.000Z (about 12 years ago)
• Last Synced: 2023-04-10T14:22:17.848Z (over 1 year ago)
• Language: C
• Size: 160 KB
• Stars: 2
• Watchers: 4
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# nagios2graphite

`nagios2graphite` is a nagios event broker module that provides nagios information on an UDP graphite server

** Still in Alpha version - and does not do what the README say - but close - Hack away **

currently supports service and host checks

## Requirements

Configure graphite with UDP enabled. In your `carbon.cfg`:

    ENABLE_UDP_LISTENER = True

    UDP_RECEIVER_INTERFACE = 127.0.0.1

    UDP_RECEIVER_PORT = 2003

Why UDP? We don't want our monitoring system to become unstable if our graphite goes down do we?

## Compilation

Requires basic compile tools to be installed.

    make

    make install

## Configuration

Configure broker events and add the broker module to your `nagios.cfg`.

    event_broker_options=-1 # Yes that's -1

    broker_module=/var/lib/nagios3/nagios2graphite.o

This will send to localhost on port 2003. To change this supply:

    broker_module=/var/lib/nagios3/nagios2graphite.o host="somehost.org" port=2003

This will send to `somehost.org` on port 2003. To activate debug:

    broker_module=/var/lib/nagios3/nagios2graphite.o host="somehost.org" port=2003 debug=1

Note: If you already have a broker_module line, you need to ADD another line. Nagios will define multiple brokers active.

## Loading

You need to restart your nagios config if you changed this lines.

Check in `/var/log/nagios3/nagios.log` to see if the module is loaded

    [1348838041] graphite: nagios2graphite by Patrick Debois

    [1348838041] graphite: successfully finished initialization

    [1348838041] Event broker module '/var/lib/nagios3/nagios2graphite.o' initialized successfully.

## Sample debug output

    [1348836352] graphite: nagios.vagrant-client.servicecheck.free_memory.current_attempt 3 1348836352

    [1348836352] graphite: nagios.vagrant-client.servicecheck.free_memory.max_attempts 3 1348836352

    [1348836352] graphite: nagios.vagrant-client.servicecheck.free_memory.state_type 1 1348836352

    [1348836352] graphite: nagios.vagrant-client.servicecheck.free_memory.state 2 1348836352

    [1348836352] graphite: nagios.vagrant-client.servicecheck.free_memory.execution_time 1.002262 1348836352

    [1348836352] graphite: nagios.vagrant-client.servicecheck.free_memory.early_timeout 0 1348836352

    [1348836352] graphite: nagios.vagrant-client.servicecheck.free_memory.latency 0.243000 1348836352

    [1348836352] graphite: nagios.vagrant-client.servicecheck.free_memory.return_code 2 1348836352

## Todo

- more robust error handling (hey my C is rusty :)

- expose more internal information: f.i. NOTIFICATION-DATA, DOWNTIME-DATA, FLAPPING-DATA, EXTERNAL-COMMAND-DATA, CONTACT-NOTIFICATION-DATA - See 

- Correctly handle metrics that change their metric size (us,ms,s) or (mb,gb,tb) : See 

## Alternatives

- Graphios: python based -  to graphite

- Gearphite: python based -  to opentsdb

- Icinga-to-graphite: 

- Via collectd:  and then you can forward it to graphite using carbon-writer

- @auxesis on latency in checks -