Backend

The backend system has three major components:

REST API Server - Flask-based web server providing HTTP REST endpoints for system configuration and status
Monitor Service - Core monitoring daemon handling hardware I/O, real-time clock, GSM communication, and system state management
PostgreSQL Database - Persistent storage for configurations, user data, alerts, and system logs

Source code¶

The source code of the backend system: arpi_server. It is a sub-module of the arpi_management project: arpi_management

You find how to get the source code here!

Architecture¶

erDiagram
  NGINX }|--|{ WEBAPP-SOURCE : "static files"
  NGINX }|--|| SERVER : "REST API"
  NGINX }|--|| MONITOR : SOCKET-IO
  SERVER ||--|| MONITOR : IPC-SOCKET
  SERVER {
    type flask
  }
  MONITOR }|--|| DATABASE : psycopg2
  MONITOR {
    type python-threading
    server socket-io
  }
  SERVER }|--|| DATABASE : psycopg2
  DATABASE {
    type postgresql
  }

Port Configuration¶

Connection	Development	Production
NGINX communication	4200	80, 443
SERVER REST API	8080	file socket
MONITOR socket-io	8081	8081
DATABASE psycopg2	5432	5432

Preparing the database for development¶

When you want to run the system on your local machine you have to prepare the database content first. The database needs a password for the service user (called argus). You have to update the value for DB_PASSWORD in server/etc/secrets.env.

# start the database
./scripts/start_database.sh
# prepare the users and tables
pipenv run flask init-db
pipenv run flask migrate
pipenv run flask upgrade
# prepare a configuration
pipenv run src/data.py -d -c test_01

Starting the backend services in development mode¶

You can run the backend services in development mode locally with mock adapters.

# go to the server folder
cd server
# start the database
./scripts/start_database.sh
# start the REST API
pipenv run flask run
# or
pipenv run start-server

# in another terminal start the monitoring service
pipenv run python -d -s -m monitor
# or
pipenv run start-monitor

REST API is available on: http://localhost:8080

Deploying the backend services to Raspberry¶

# go to the arpi_management project
pipenv run ./install.py -vpe prod server
pipenv run ./install.py -vpe prod monitor

Starting the backend services in production mode¶

On the Raspberry PI the backend services can be managed as systemd services.

sudo systemctl start argus_server
sudo systemctl start argus_monitor