esp-prom

Run a Prometheus monitoring endpoint on an ESP32 or other Espressif microcontroller. Increase obervability in your project while you develop, by monitoring heap memory, FreeRTOS task stack size, and other helpful metrics. For use with the Espressif IoT Development Framework (ESP-IDF).

This is an implementation of digitalocean/prometheus-client-c, with esp_http_server from ESP-IDF to serve the endpoint.

Usage

Clone this repository, e.g.:

cd ~/myprojects/esp
git clone https://github.com/nated0g/esp-prom

Add path to components in CMakeLists.txt:

cmake_minimum_required(VERSION 3.5)
set(EXTRA_COMPONENT_DIRS /home/user/myprojects/esp/esp-prom/components)
include($ENV{IDF_PATH}/tools/cmake/project.cmake)
project(my-esp-project)

Example

Check the /examples directory for a simple example using the default metrics you can flash to an ESP32.

Configure project with idf.py menuconfig to add WiFi credentials, then build and flash. View the logs to get the IP address of the device. You'll need this to set up a Prometheus scraper.

cd ~/myprojects/esp/esp-prom/examples/simple
idf.py menuconfig
idf.py build
idf.py flash
idf.py monitor

Make note of the IP address of the device.

Prometheus/Grafana Setup

Included in this repository is a simple Prometheus/Grafana setup in a docker-compose.yml file, along with some neccesary configuration files.

Installing Docker is outside of the scope of this document, you can learn about that here: Get Docker.

Edit docker/prometheus/prometheus.yml with the IP address of your device.

...
  - job_name: 'esp32'
    
    static_configs:
      - targets: ['192.168.1.178']

Start the containers with docker-compose up.

Prometheus will be served on localhost:9090, and Grafana will be served on localhost:3000.

Set up some dashboards in Grafana and enjoy the added visibility into your firmware!

To do:

• Add a default dashboard to Grafana
• Add docs for adding user defined metrics
• Add more default metrics