Node-RED is a visual programming tool that allows users to create applications by connecting "nodes" together. It is often used for Internet of Things (IoT) applications but can be used for a wide variety of purposes. It is built on top of Node.js, a JavaScript runtime environment, and uses a flow-based programming model, which means that applications are created by connecting together small units of functionality, or "nodes," in a visual manner.

Node-RED installation

First we installed Git, Node.js and npm on the OpenWRT Router with the opkg package manager

Node-RED installation via NPM was not successful with the following command due to dependency issues.

npm install node-red

Hence the source code for Node-RED was cloned from (https://github.com/node-red/node-red) and the following commands were used to install and build the code.

1. npm install (install packages)
2. npm run build (build root)
3. npm start (start the node-red sever)

With the above process, we were able to run Node-RED at localhost:1880 on the OpenWRT Raspberry Pi.

Starting Node-RED on Bootup

This process can vary based on the init system, on OpenWRT you can follow the following steps.

1. Create a script called 'S99StartNodeRed' and put the following lines in the script.

#!/bin/sh

/path/to/node-red/folder

npm start

2. Give the execution permission by running the following command.

chmod +x S99StartNodeRed

Alternatively, you can also copy the Node-RED binary which is built to the /usr/bin directory and execute it from an init script.

Setting up Node-RED to communicate over TCP

The nodes to communicate over TCP are available by default and can be used to listen or connect to a TCP socket on a specific port of a host on the network. There are specific nodes to receive or send data. When data is fed into the input of a TCP out node, the data is sent over TCP to the connected.

Additional nodes can be installed via the 'Manage Palette' sub-menu. In the below image, the 'node-red-dashboard' (installed via palette) is a collection of nodes with which a functional dashboard can be created.

Note: Internet Access is needed on the OpenWRT Router to install new nodes into Node-RED.

Using Node-RED on OpenWRT to Manage Applications on the Network

Node-RED on OpenWRT can be used to set up a Management Server to send commands to the Applications running on the network. The Applications on the network will need to implement a socket to connect to the management server.

It is preferred to run the management socket on a thread separate from the application code. An implementation for a project is explained below with links to relevant pieces of code.

Management Interface with TCP Socket

• In this setup, there is a UI with control elements to manage the demo and features. This control/management data is sent over to the client and server over TCP sockets. The client and server implement a secondary management socket on a different thread to receive these messages and act on the main thread which is running the demo. The above diagram shows the implementation.
• The code to set up a management socket and parse simple commands can be found here in the main function.

The specific Node-RED config file and instructions to use it can be found in the issue tracker below. The config file is a JSON structure describing different nodes present and their configurations too.

The issue tracker also contains screen captures of the Management User Interface.

The issue tracker -> issue