johnmmason / pimonitor

piMonitor is a simple application for recording, storing, and sharing sensor data.

Project Overview

Consumer grade IOT sensors offer convenience and peace-of mind, but lack features important to tech hobbyists such as

• Sensor flexibility,
• customization,
• data security,
• and affordability.

This project attempts to address these issues by creating a software framework for inexpensive, homemade Raspberry-Pi based sensor modules and an accompanying self-hosted data logging server.

This project is still early in its development, so its functionality is rather limited. Currently, the system supports a DHT22 temperature sensor but could be easily modified for other similar sensors. Please see below for a list of planned additions to the code.

Project Components

This project consists of two main components, the server and client(s). This repository contains only the server code. Corresponding client code can be found in johnmmason/pimonitor-sensor

The Server

The server can either be hosted locally or on a cloud instance such as AWS.

The server consists of a Flask-based WSGI application which accepts JSON data (later referred to as "the api") from the sensors and stores the data in a PostgreSQL database.

The Client

The client code was developed to run on a Raspberry Pi, and has been tested on both the Raspberry Pi 4 and Raspberry Pi Zero.

The client script reads data from attached sensors (currently supported, a DHT22 temperature and humidity sensor), compiles data into a JSON string, and then sends it to the API via a HTTP POST request.

Setup and Installation

Database Setup

You will need the following

• PostgreSQL with a database for this project (this can, but does not have to be, the default database)
• A user with adequate permissions to perform CRUD operations on the project database (this can, but does not have to, be the default user)
• Appropriate configuration to allow remote connections to your database

Open a SQL prompt using a method of your choice and create a table for this project:

CREATE TABLE home_data(
    id SERIAL,
    location VARCHAR(30),
    timestamp TIMESTAMP,
    temperature REAL,
    humidity real
)

Server Setup

First, update your system and install required dependencies:

sudo apt-get update && sudo apt-get upgrade -y
sudo apt-get install python3-dev python3-pip python3-venv -y

Clone this repository:

git clone https://github.com/johnmmason/pimonitor.git

Navigate to the project directory:

cd pimonitor
pwd

Create a Python virtual environment and install Python packages:

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

Create sample_config.ini and configure database settings:

cp sample_config.ini config.ini
emacs config.ini

Set environment variables for Flask:

export FLASK_APP=pimonitor
export FLASK_ENV=development

Run the app using the built-in Flask development server:

flask run --host=0.0.0.0

Now, configure a production WSGI server to serve the application. piMonitor has been tested with Gunicorn.

Features Coming Soon

• Split server code into modules to allow for easy modification and customization
• Add data monitoring functionality to allow for alerts when sensor values cross a predefined threshold
• Add email reporting functionality to share sensor data at a predefined interval
• Add support for multiple sensors on one client device
• Add native support for additional types of sensors