IoT - logger 🤖

IoT logger is an IoT architecture for gathering informations about IoT sensors. It is based on RabbitMQ queues and works with Nuclio.

What is RabbitMQ

From the RabbitMQ home page:

RabbitMQ is the most widely deployed open source message broker.

What is Nuclio

From the Nuclio home page:

Nuclio is an open source and managed serverless platform used to minimize development and maintenance overhead and automate the deployment of data-science based applications

Architecture

The architecture is composed by some sensors that send informations about their status on a queue (iot/web/sensors). When a new message is incoming on this queue a nuclio function is triggered. This function send an HTTP request to a middleware written in python (http://<service_url>:<service_port>/iot/logger/sendlog).
The middleware is in communication with a UI web based made in React.js by a socket-io connection.
With this GUI we can read the status of all sensors connected and we can change their status. When we send a message from GUI to middleware this message will be send on another queue (iot/sensor/{sensor_id}).
Each sensor is connected to their queue by own sensor_id.

🚀 Installation

The project is composed of 5 parts

RabbitMQ
Nuclio
front-end (React.js)
middleware (Python)
IoT sensors (Python)

You can install this components one by one with dockerfile, or you can use a docker-compose.yml file provided. First of all clone the project and follow the instructions below

git clone https://github.com/mariomamo/iot-project

⚙️ Some configuration before starting

Before starting you need to modify some configurations files.
1️⃣ Change IP address in front-end web app Go to \iot-logger\iot-front-end\compiled and edit index.html file. You can find a div with "root" id, you need to change the IP address in url property with yours.

<div id="root" url="http://<INSERT_YOUR_LOCAL_IP_ADDRESS_HERE>" port="5000"></div>

If you want you can change also the port, but it is useless, you can map this port on another port of docker.

2️⃣ Change the IP addresses in application.yaml inside python files Go to iot-logger\iot-logger and edit application.yml file. You have to write your local IP address in topic_url property.

name: 'IoT-Logger'  
socket_io_port: 5000  
receive_topic_name: 'sensors/logs'  
send_topic_name: 'iot/sensors'  
topic_url: '192.168.1.59'  
rest_port: 4444

⚠️ If you want to change the port remember to change socket_io_port also ⚠️
⚠️ rest_port is the port used for REST API. ⚠️
⚠️ If you want to change this port remember to change the port inside nuclio script also ⚠️

Go to iot-logger\iot-sensors and edit application.yml file.

sensors:  
  - chat_id: "1"  
    name: 'Car'  
    sensor_type: "car"  
    image: 'https://upload.wikimedia.org/wikipedia/commons/thumb/6/65/Circle-icons-car.svg/1200px-Circle-icons-car.svg.png'  
    receive_topic_name: 'iot/sensors'  
    send_topic_name: 'iot/web/sensors'  
    topic_url: '192.168.1.59'  
  - chat_id: "2"  
    name: 'My house'  
    sensor_type: "home"  
    image: 'https://media.istockphoto.com/vectors/smart-home-vector-icon-with-airwaves-isolated-on-white-background-vector-id1186712143?k=6&m=1186712143&s=170667a&w=0&h=zZoIDzrMKXkUERFYeuSnYhiuSxZ22IcXT0RPIanZxG0='  
    receive_topic_name: 'iot/sensors'  
    send_topic_name: 'iot/web/sensors'  
    topic_url: '192.168.1.59'  
  - chat_id: "3"  
    name: 'Bed room air conditioner'  
    sensor_type: "air-conditioner"  
    image: 'https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcRfe3ft1wqiQUu1Kra38EezbmQTljHkFPSPzA&usqp=CAU'  
    receive_topic_name: 'iot/sensors'  
    send_topic_name: 'iot/web/sensors'  
    topic_url: '192.168.1.59'

You can simply add a new sensor adding a new element in the sensors array. You just have to add the required informations. Actually there are only three types of sensors supported

car
home
air-conditioner

Other sensor types will be ignored. Also in this case you need to change the current IP in topic_url with yours. Each sensor can have different topic_url.

⚠️ You can of course change all the topics, but remember to keep these topics name coherents ⚠️

🐳 docker-compose

you just have to type these commands

docker-compose up --build -d

build: rebuild images if they are already builded before
d: execute in background. When the docker-compose will setted up you will not see the docker logs console

After this operation some containers (iot-logger and iot-logger) can fail. This is because rabbitMQ container is not ready. You can simply avoid this by restarting failing containers.

You can skip the next section if you have run docker-compose.

🐋 dockerfile

1️⃣ RabbitMQ

In some cases you need to deploy only some parts of this projects for load balancing or availability purpose. So you can also use dockerfile instead of docker-compose.

Open your terminal and copy/paste this line

docker run -p 9000:15672 -p 1883:1883 -p 5672:5672 -d --name rabbitMQ cyrilix/rabbitmq-mqtt

✔️ It is also available for ARM processors

docker run -p 9000:15672 -p 1883:1883 -p 5672:5672 -d --name rabbitMQ arm32v7/rabbitmq

2️⃣ Nuclio

Open your terminal and copy/paste this line

docker run -p 8070:8070 -v /var/run/docker.sock:/var/run/docker.sock -v /tmp:/tmp -d --name nuclio nuclio/dashboard:stable-amd64

❌ It is not available for ARM processors

3️⃣ IoT logger

Go to iot-logger folder :warning: Check the directory where you are :warning:

cd iot-logger

Build the image

docker build -t iot-logger .

Run container

docker run -v <path_where_you_cloned_project>\iot-logger\iot-logger:/logger -dp 5000:5000 -dp 4444:4444 --name iot-logger -it iot-logger

4️⃣ IoT sensors

Go to iot-sensors folder :warning: Check the directory where you are :warning:

cd iot-sensors

Build the image

docker build -t iot-sensors .

Run container

docker run -v <path_where_you_cloned_project>\iot-logger\iot-sensors:/sensors -d --name iot-sensors -it iot-sensors

5️⃣ IoT frontend

Go to iot-front-end folder :warning: Check the directory where you are :warning:

cd iot-front-end

In questo caso non abbiamo bisogno di un dockerfile Run docker

docker run -v <path_where_you_cloned_project>\iot-logger\iot-front-end\compiled:/usr/local/apache2/htdocs/ -e TZ=Europe/Rome -dp 80:80 --name iot-frontend httpd:2.4

⚙️ Nuclio configuration

Once you have installed the project you need to configure it. Go to iot-logger/nuclio-function/loggerconsumer.yaml and write your local ip address at line 17

triggers:
    MyMQTTTriggher:
      class: ""
      kind: rabbit-mq
      url: "amqp://guest:guest@<INSERT_YOUR_LOCAL_IP_ADDRESS_HERE>:5672"
      attributes:
        exchangeName: "\"\""
        queueName: iot/web/sensors
        topics:
          - '""'

Open Nuclio control panel, create a new project and import the function you have just modified. In the next window you should change the existing IP addres at line 25 with yours.

response = requests.get("http://<INSERT_YOUR_LOCAL_IP_ADDRESS_HERE>:4444/iot/logger/sendlog", headers=headers, json=payload)

Now you can deploy your function. When your function has been deployed go to IoT logger webapp and restart iot-logger container. When iot-logger will be restarted it will send some information to the iot-logger and you will se these informations on the webapp. Now you can send commands to devices and you can read their responses. Enjoy :)

⚠️ Important

Every time you restart nuclio you need to re-deploy nuclio function

The images in this repository were taken from flaticon.com

mariomamo/iot-logger