Brew Dashboard

Full solution to monitor a temparature sensor connected to a Raspberry Pi. Includes a React.js web with near real-time temperature chart and Azure Functions to process temperature readings and alert system

Architecture

This solution has been created using a PolyCloud approach in which every component is hosted in a different cloud provider.

For the RaspberryPi, py-temperature-recorder has been used to send data to the Azure Function.

Content structure

In this repository you will find all components needed to make this solution work.

Web

The /web folder contains all the resources for the React Web App. The easiest way to start it is by running the following command from the root folder (/web's parent folder):

yarn start

This will get into the /web folder, install dependencies and start the server.

Alternatively you can run the following commands in the /web folder:

yarn
yarn start

To create a production ready version of the web app run:

yarn build

If you don't have yarn install it following the steps in Yarn official site.

Functions

In /functions folder there are two serverless functions:

• /setAndCheckTemperature: this function is meant to receive temperature readings, store it through a GraphQL endpoint and verify if measure value is outside the temperature thresholds the user has define in settings.
• /bot: the serverless function that act as bot endpoint to process requests (made for Telegram, some other chat clients might require modifications).

setAndCheckTemperature

The function /setAndCheckTemperature is made with NodeJS and Serverless Framework to be deployed in Azure. To deploy it you first need to install serverless npm package globally:

npm i -g serverless

Since this function is an Azure Function, the serverless-azure-functions plugin is also needed:

npm i -g serverless-azure-functions

Once installed follow this steps to setup your Serverless environment and Azure credentials, then run:

serverless deploy

This command will use serverless.yml to package, provision and deploy your function into the selected Azure subscription.

To make the funciton work three settings have to be defined in web.config in the Azure portal:

• GraphQlUrl: The GraphQL endpoint URL (when using AWS AppSync the API endpoint showed in the dashboard).
• GraphQlKey: The API Key to authorize calling the GraphQL endpoint.
• NotificationsUrl: The endpoint to send user and measure data when is outside the thresholds in user settings.

When deployed, the URL generated is the one the temperature sensor have to call with measure data.

Bot

TBD

GraphQL

To retrieve and store the data this solution uses GraphQL. The schema and resolvers found in /graphql folder contains the schema and resolvers (written in VTL) used in AWS AppSync connected to an AWS DynamoDB.