Weather Edge Configuration

Clone this repository to your edge device (Raspberry Pi, Ubuntu, Debian are supported)

git clone https://github.com/Itea-Lab/Weather-Edge.git
cd Weather-Edge

Initial Setup

1. Create Environment Configuration File

Create a .env file in the project root to store all necessary credentials:

sudo nano .env

2. Environment Configuration Template

# InfluxDB Configuration
INFLUXDB_USERNAME=<username>
INFLUXDB_PASSWORD=your_password
INFLUXDB_ORG=weather_org
INFLUXDB_BUCKET=weather_data
# For Docker containers, use service name
INFLUXDB_ROUTE=http://database:8086
# Token will be obtained after InfluxDB initialization (see step 6)
INFLUXDB_TOKEN=<your_token_here>

# Data Configuration
DATA_LOCATION=<your_location_name> (district1, district2, etc)
MEASUREMENT_NAME=weather_sensor

# MQTT Configuration
MQTT_PUB=espclient
SUB_USERNAME=raspiclient
MQTT_PASSWORD=your_mqtt_password
MQTT_TOPIC=weather/data
MQTT_PORT=1883
BROKER_ENDPOINT=mosquitto

# Dashboard Configuration
JWT_SECRET=your-super-secret-jwt-key-here-make-it-long-and-random
JWT_EXPIRES_IN=1d
ADMIN_USERNAME=admin
ADMIN_NAME=Administrator
ADMIN_PASSWORD_HASH=your_password_hash_here
ADMIN_EMAIL=admin@example.com
ADMIN_ROLE=admin
TEST_USERNAME=testuser
TEST_NAME=Test User
TEST_PASSWORD_HASH=your_password_hash_here
TEST_EMAIL=test@example.com
TEST_ROLE=user

3. Project Structure

weather-edge-config/
├── .env                       # Main environment variables
├── docker-compose.yml         # Docker services configuration
├── README.md
├── mosquitto-init/            # MQTT broker initialization
│   └── entrypoint.sh
├── mosquitto-config/          # MQTT broker config (auto-generated)
├── influxdb2-config/          # InfluxDB config (auto-generated)
├── influxdb2-data/            # InfluxDB data (auto-generated)
└── data-processor/
    ├── .env                   # Processor-specific env FOR LOCAL TESTING (copy of main .env)
    ├── credentials/           # Device certificates and keys (See step 4)
    ├── Dockerfile
    ├── requirements.txt
    ├── run.py                 # Application entry point
    ├── dataProcessor.py       # Main processing controller
    ├── config/
    │   ├── __init__.py
    │   └── config.py          # Centralized configuration
    ├── mqtt/
    │   ├── mqttReceiver.py    # MQTT message receiver
    │   └── mqttPublisher.py   # MQTT message publisher
    └── db/
        ├── influxClient.py    # InfluxDB connection management
        └── influxWriter.py    # Data writing operations

Deployment Steps

4. Device Registration and AWS IoT Certificates

Before running the weather edge processor, you must register your device with the Weather Platform to obtain AWS IoT certificates:

4.1. Device Registration Process

1. Contact Weather Platform Administrator or access the device registration portal

2. Provide Device Information:

   • Device name/identifier (e.g., raspi-1, weather-station-01)
   • Location/district assignment (e.g., district5, station_01)
   • Device type: Weather Edge Processor

3. Receive Certificate Package: You will receive a package containing:

   • {device-name}_certificate.pem - Device certificate
   • {device-name}_private_key.pem - Private key
   • {device-name}_public_key.pem - Public key (optional)
   • {device-name}_connection_info.json - Connection configuration
   • AmazonRootCA1.pem - Root CA certificate

4.2. Certificate Installation

1. Create credentials directory:

   mkdir -p data-processor/credentials

2. Copy certificate files: Place all received certificate files in data-processor/credentials/:

   data-processor/credentials/
   ├── raspi-1_certificate.pem
   ├── raspi-1_private_key.pem
   ├── raspi-1_public_key.pem
   ├── raspi-1_connection_info.json
   └── AmazonRootCA1.pem
   

3. Verify connection info format: The connection_info.json should contain:

   {
     "endpoint": "your-iot-endpoint.iot.us-east-1.amazonaws.com",
     "port": 8883,
     "topics": {
       "publish": "weatherPlatform/telemetry"
     }
   }

4. Set proper permissions:

   chmod 600 data-processor/credentials/*_private_key.pem
   chmod 644 data-processor/credentials/*.pem
   chmod 644 data-processor/credentials/*.json

4.3. Configuration Verification

• Ensure DATA_LOCATION in your .env file matches your registered location
• The system will automatically discover certificate files based on naming patterns
• Dynamic topic routing will append your location to create: weatherPlatform/telemetry/{DATA_LOCATION}

5. Initial Docker Compose Setup

Enable execute permission for script file

chmod +x mosquitto-init/entrypoint.sh

Start the services for the first time:

docker compose up --build -d

6. InfluxDB Token Setup (IMPORTANT)

After the initial startup, you need to obtain the InfluxDB token:

1. Access InfluxDB Web Interface:

   http://localhost:8086
   

2. Initial Setup:

   • Username: Use INFLUXDB_USERNAME from your .env
   • Password: Use INFLUXDB_PASSWORD from your .env
   • Organization: Use INFLUXDB_ORG from your .env
   • Bucket: Use INFLUXDB_BUCKET from your .env

3. Get API Token:

   • Navigate to "Data" → "API Tokens"
   • Click "Generate API Token" → "All Access API Token"
   • Copy the generated token

4. Update Environment Files: Update both .env files with the token:

   INFLUXDB_TOKEN=your_copied_token_here

5. Restart Services:

   docker compose down
   docker compose up -d

7. MQTT Broker Configuration

The MQTT broker supports two authentication modes:

Option A: Anonymous Access (Development)

Edit mosquitto-config/mosquitto.conf and set:

allow_anonymous true

Option B: User Authentication (Production - Recommended)

The system automatically creates authentication files:

• mosquitto-config/passwd - Stores username/password pairs
• mosquitto-config/mosquitto.acl - Access control list

Default Users Created:

• espclient - Can publish (write) to all topics
• raspiclient - Can subscribe (read) from all topics

Testing MQTT Connection:

Without authentication:

mosquitto_sub -h localhost -p 1883 -t weather/data
# Returns: Connection error: Connection Refused: not authorised.

With authentication:

mosquitto_sub -h localhost -p 1883 -u raspiclient -P your_mqtt_password -t weather/data

8. Running the Weather Data Processor

The data processor automatically:

• Connects to InfluxDB using the configured token
• Subscribes to MQTT topics for weather data
• Processes and stores incoming data
• Validates data format and handles errors

Manual Testing:

cd data-processor
python run.py

Production (Docker): The processor runs automatically as part of the Docker Compose stack.

Accessing Services

• InfluxDB UI: http://localhost:8086
• Dashboard: http://localhost:3000 (if enabled)
• MQTT Broker: localhost:1883

Configuration Variables Reference

Variable	Purpose	Example
INFLUXDB_TOKEN	Database API access	eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9...
MQTT_PUB	MQTT publisher username	espclient
SUB_USERNAME	MQTT subscriber username	raspiclient
BROKER_ENDPOINT	MQTT broker address	Raspi's IP address
DATA_LOCATION	Sensor location tag	district5, station_01
MEASUREMENT_NAME	InfluxDB measurement name	weather_sensor

Troubleshooting

Common Issues:

1. AWS IoT Certificate Issues:

   • Missing certificates: Ensure all certificate files are in data-processor/credentials/
   • Permission denied: Check file permissions (private keys should be 600)
   • Connection refused: Verify device is registered and certificates are valid
   • Topic permission denied: Ensure AWS IoT policy allows publishing to weatherPlatform/telemetry/*

2. "Not authorized" MQTT error:

   • Check username/password in .env
   • Verify ACL permissions
   • Ensure user exists in passwd file

3. InfluxDB connection failed:

   • Verify INFLUXDB_TOKEN is correct
   • Check if InfluxDB service is running
   • Validate organization and bucket names

4. Environment variables not loading:

   • Ensure .env file exists in correct location
   • Check for typos in variable names
   • Restart containers after changes

Debug Commands:

# Check container status
docker compose ps

# View logs
docker compose logs weather-edge-processor
docker compose logs database
docker compose logs mosquitto

# Test MQTT connectivity
docker exec -it mqtt-broker mosquitto_sub -h localhost -u raspiclient -P your_password -t weather/data

# Test AWS IoT certificate discovery
cd data-processor
python -c "from mqtt.mqttPublisher import AWSIoTPublisher; pub = AWSIoTPublisher(); print('Certificates discovered successfully')"

# Verify certificate files
ls -la data-processor/credentials/