The GMNI Project

The GMNI project is just a hobby project to learn a little bit more about IoT devices and their use in the home. To start, I'm using an ESP8266 to collect temperature and humidity measurements from a DHT22 sensor, then sending that information to a central server to be stored and graphed. For collecting, storing and graphing I'm using what I dubbed the GMNI stack:

• Grafana - Used to build a Dashboard to display all collected data from sensors
• Mosquitto - "A message broker that implements the MQTT broker". The IoT devices will publish their sensors data to this message broker, which is responsible to deliver the message to the appropriate subscriber(s).
• Node-Red - "A Visual Tool for wiring the Internet of Things". Node-Red allows me to graphically create a subscriber of the mqtt broker, massage the data and send it to InfluxDB to be stored in the correct format.
• InfluxDB - A time series database. InfluxDB is used to store the data from the sensors. It will then serve as the data source for Grafana.

This repo contains ansible scripts to install the GMNI stack on a server. It currently supports CentOS and Fedora (It is also very close to working on a Raspberry Pi with Fedora 25).

The Vagrantfile allows you to test on a virtual machine, to get started simply run vagrant up and stack will be configured for you, the ports for each service are printed at the end of the ansible output.

The temperature_sensor directory contains the code used to run on the ESP8266 devices. The sample directory contains sample data to test loading the data into InfluxDB. It was usefult to start playing with Grafana dashboards.

In the future, I'd like to provide some Dashboard templates.

Let me know what you think and patches are welcome.

How to get started:

• vagrant up
• Use mosquitto_pub to publish data to the mqtt broker or use script in sample dir to load some sample temperatures directly to InfluxDB.
  • mosquitto_pub -d -t sensors/testroom/temp -m '72'
• Using the debug panel, you can also see the data on node-red.
• Start building a Dashboard on Grafana to visualize data.