Getting Started

To get started with prometheus-virtual-metrics, you can use the example project as a starting point or follow this tutorial.

Installation

$ pip install prometheus-virtual-metrics

Starting The Server

prometheus-virtual-metrics is started by invoking python3 -m prometheus_virtual_metrics. All configuration is done by specifying a settings module by either setting the environment variable PROMETHEUS_VIRTUAL_METRICS_SETTINGS or the command line flag -s.

The settings module needs to be specified as a Python import string (for example: my_project.settings, not my_project/settings.py) and has to be importable. Documentation on all available settings values can be found here.

prometheus-virtual-metrics can be used with watchfiles.

Examples:

# command line flag
python3 -m prometheus_virtual_metrics -s my_project.settings

# environment variable
PROMETHEUS_VIRTUAL_METRICS_SETTINGS=my_project.settings python3 -m prometheus_virtual_metrics

# watchfiles
python3 -m watchfiles \
    "python3 -m prometheus_virtual_metrics -s my_project.settings" \
    /app/my_project/

Connecting Grafana

Once the server runs, you can open Grafana and connect to prometheus-virtual-metrics like you would connect to a regular Prometheus server: Link

The prometheus-virtual-metrics default settings configure an example plugin that exposes a metric called prometheus_virtual_metrics_example_metric. To verify that everything works, you can go to Grafanas explore view and query prometheus_virtual_metrics_example_metric.

Write A Plugin

Write a Plugin and put it somewhere your settings module can import it. Then, load the plugin into your settings module and add it to PLUGINS:

# settings.py
from my_project.plugins import MyPlugin

PLUGINS = [
    MyPlugin(),
]