paolobasso99/uptimerobot_exporter

A Prometheus exporter for uptimerobot.com

uptimerobot_exporter

uptimerobot_exporter is a Prometheus exporter for Uptimerobot.

Setup

Exporter

The easier way to setup this exporter is to use the Docker image on Docker Hub. Using a docker-compose.yaml:

version: "3"

services:
uptimerobot_exporter:
    image: paolobasso/uptimerobot_exporter
    container_name: uptimerobot_exporter
    environment:
    - UPTIMEROBOT_READ_API_KEY=your-api-key
    - INTERVAL_SECONDS=300
    - LOG_LEVEL=INFO
    - PORT=8000
    ports:
    - 8000:8000
    restart: unless-stopped

If Prometheus is in the same server an internal Docker network without exposing the metrics to the internet would be preferred.

Envirorment variables

Variable	Required	Default	Description
UPTIMEROBOT_READ_API_KEY	YES		Your Uptimerobot read API key. Found on the Uptimerobot's My Settings page -> API Settings.
INTERVAL_SECONDS	NO	300	How many seconds to wait between a scrape end and the next scrape. You should use the Uptimerobot monitor's shortest interval.
LOG_LEVEL	NO	INFO	The log level.
PORT	NO	8000	The port where metrics will be exposed

Prometheus

Add a job to your Prometheus configs:

- job_name: 'uptimerobot'
   scrape_interval: 5m
   scrape_timeout: 300s # Same as INTERVAL_SECONDS env vars
   static_configs:
     - targets: 
        - 'localhost:8000' # Use the PORT env var (Default is 8000)

Metrics exposed

Metric	Type	Labels	Description
uptimerobot_up	Gauge		The last scrape was successful
uptimerobot_scrape_duration_milliseconds	Gauge		The duration of the last scrape in seconds
uptimerobot_monitor_status	Gauge	id, url, name, type	Status of the monitor: 0 = paused, 1 = not checked, 2 = up, 8 = seems down, 9 = down
uptimerobot_monitor_response_time_millisecond	Gauge	id, url, name, type, status	Last response time of the monitor in milliseconds
uptimerobot_monitor_response_time_average_milliseconds	Gauge	id, url, name, type	Average response time of the monitor in milliseconds
uptimerobot_monitor_log_type	Gauge	id, url, name, type	Last log type of the monitor: 1 = down, 2 = up, 98 = started, 99 = paused
uptimerobot_monitor_log_datetime	Gauge	id, url, name, type, logtype	Last log of the monitor datetime

Grafana Dashboard

You can find the Grafana dashboard to visualize the metrics exposed by this exporter here and it looks like this:

Why

I needed a Prometheus exporter for Uptimerobot and the existing ones that I found are either old, not updated, poorly documented or they expose not enough metrics. It was also an opportunity to learn more about Prometheus and Python.

Technologies

1. Prometheus
2. Uptimerobot
3. Grafana
4. Python
5. Docker
6. GitHub Actions

Things I learnt

1. Using Uptimerobot to monitor websites uptime
2. Creating a basic Prometheus Exporter with Python
3. Creating a Grafana Dashboard from scratch
4. Dockerizing a simple Python application
5. Auto publishing Docker images to Docker Hub with GitHub actions
6. Tesing a simple python application
7. Self Hosting Prometheus
8. Self Hosting Grafana
9. Python Docstrings

License

GNU AFFERO GENERAL PUBLIC LICENSE Version 3