Plugins & custom checks

Lookout watches CPU, memory, disk, and services out of the box. Plugins let you check anything else — a website is up, a certificate isn’t expiring, a queue isn’t backing up — in any language you like.

Status: the plugin runner is part of the next phase. This page documents the model so you can write checks that will drop straight in.

Nagios-compatible

If you’ve used Nagios, you already know how Lookout plugins work. A plugin is any program that:

• prints a human-readable status line, and
• exits with a code that means the result.

This means the thousands of existing Nagios plugins work with Lookout unchanged.

A check in any language

Here’s a complete check in shell — “is the website returning 200?”:

#!/usr/bin/env bash
code=$(curl -s -o /dev/null -w "%{http_code}" https://example.com)
if [ "$code" = "200" ]; then
  echo "OK - site is up ($code)"; exit 0
else
  echo "CRITICAL - site returned $code"; exit 2
fi

Performance data

Append metrics after a | so Lookout can graph them over time:

echo "OK - response 142ms | response_ms=142 status=200"; exit 0

Installing a custom plugin

Drop the script in the agent’s plugins folder with a small manifest describing how often to run it:

# /etc/lookout/plugins/site-up.yaml
name: site-up
command: ./site-up.sh
interval: 60s

The agent runs it on schedule and reports the result to the dashboard, where it shows up like any other check and can trigger alerts.