pgflux

pgflux is a Python utility to send PostgreSQL metrics to InfluxDB. The drivers when developing this tool were:

• Make it easy to extend: Authoring new queries should only require PostgreSQL knowledge to allow non-devs to easily add/fix/modify the collected metrics. pgflux does not require you to know Python, InfluxDB nor Grafana to work on the provided SQL queries.
• Be independently runnable: pgflux should be runnable without first setting up a hugely complex dev-environment. All that is needed is a Python environment and a PostgreSQL server (a Dockerfile for testing is supplied for the latter).

Screenshots

It enables Grafana visualisations as seen in the screenshots below:

Installation

The project can be installed directly via pip:

python3 -m venv /path/to/virtual-env
/path/to/virtual-env/bin/pip install pgflux

The use of a virtual environment is optional, but recommended.

Usage

See https://post-luxembourg.github.io/pgflux/usage.html

Development & Extending

For detailed instructions, see https://post-luxembourg.github.io/pgflux/development.html

NOTE

This project uses fabric as task runner (see https://fabfile.org). To install it use either pipx install fabric (recommended) or, if you don't have pipx: pip install --user fabric.

Commands

• Listing all available development tasks:

  fab -l

• To set up or refresh the development environment:

  fab develop

• Run PostgreSQL, InfluxDB & Grafana containers (requires docker-compose):

  fab run-dev-containers

• Run the tests:

  fab test