Skip to content

Latest commit

 

History

History
34 lines (26 loc) · 2.39 KB

README.md

File metadata and controls

34 lines (26 loc) · 2.39 KB
Raw

Sensor-api-project

Made by Troels Wittrup-Jensen

This is a project I made for monitoring tempature, humidity and air pressure using an m5go as client, and uploading the data to a raspberry pi 4 hosting a server.

It consists of 2 parts, the server(api) and client(Sensor).

Setup

To set the server up download the "api" folder and install python 3.10 and the following command for the libraries: pip install flask mariadb

Afterwards you can change the global variables in app.py to match your setup, and launch either "start-server.sh" or "start-server-debug.sh" if you would like automatic reload of the server on code-change.

To set the database up on the server download and install an sql service. I used mariadb which is an improved fork of mysql for archlinux. If you don't use arch I recommend using mysql, but it can have problems with utf8. After installing an sql-service please import the "sensor_data.sql" file using pymyadmin or the command-line.

To setup the client just download the "Sensor" folder and using platformio for vscode, simply upload the code to a m5go. Additional ECs will need to be added manually to the database.

Features

Client:

  • Monitors temperature, humidity and air pressure and displays it on it's display.
  • Shows a countdown for the next time it will try to POST the sensor-data to the server using http.
  • The middle button beneath the display can be pressed to manually trigger a POST for testing purposes.
  • LED's are present above the display to indicate it's status:
    The leftmost LED shows WiFi-status (green for connected, red for no connection).
    The LED in the middle shows status for connection to the server (green for connected, red for no connection).
    The rightmost LED blinks blue whenever it receives status status-code 200 from a POST-request it sent.

Server

  • Recieves data from clients at "http://[server-ip]:5000/api/v1/upload", and returns appropriate response codes.
  • Uses a normalised database for optimal data-storage space.
  • Only creates one connection the the database for each post request, and doesn't close it until the transaction is complete to ensure maximal performance.
  • Only proceeds with data recieves the POST-request contains a secret-key matched to one saved in the secrets table in the database.
  • Shows graphs of the sensors on "http://[server-ip]:5000/[controller-codename]" for the last two hours (can be configured in app.py).