This page contains technical information to help you get familiar with Azure IoT using the Azure IoT Starter Kit - Raspberry Pi 2. You will find two tutorials that will walk you through different scenarios: the first tutorial will show you how to connect your Azure IoT Starter kit to our Remote Monitoring preconfigured solution from Azure IoT Suite. In the second tutorial, you will leverage Azure IoT services to create your own IoT architecture.

You can choose to start with whichever tutorial you want to. If you've never worked with Azure IoT services before, we encourage you to start with the Remote Monitoring solution tutorial, because all of the Azure services will be provisioned for you in a built-in preconfigured solution. Then you can explore how each of the services work by going through the second tutorial.

We hope you enjoy the process. Please provide feedback if there's anything that we can improve.

Don't have a kit yet?: Click here

• Connecting your Raspberry Pi2 to the Azure IoT Suite Remote Monitoring Solution
• Using Microsoft Azure IoT Services to Identify Temperature Anomalies

This tutorial describes the process of taking your Microsoft Azure IoT Starter Kit for Raspberry Pi 2, and using it to develop a temperature and humidity reader that can communicate with the cloud using the Raspbian OS and Microsoft Azure IoT SDK. For Windows 10 IoT Core samples, please visit windowsondevices.com.

• 1.1 Tutorial Overview
• 1.2 Before Starting
  • 1.2.1 Required Software
  • 1.2.2 Required Hardware
• 1.3 Create a New Azure IoT Suite Remote Monitoring solution and Add Device
• 1.4 Prepare the Device
  • 1.4.1 Log in and Access the Terminal
  • 1.4.2 Log in Using PuTTY
    • 1.4.2.1 Log in using Linux or Mac OS
• 1.5 Configure the Remote Monitoring device sample
  • 1.5.1 Clone repositories
  • 1.5.2 Update the Connection String
• 1.6 Build the Modified Sample
• 1.7 View the Sensor Data from the IoT Suite Portal
• 1.8 Next steps

In this tutorial, you'll be doing the following:

• Setting up your environment on Azure using the Microsoft Azure IoT Suite Remote Monitoring preconfigured solution, getting a large portion of the set-up that would be required done in one step.
• Setting your device and sensors up so that it can communicate with both your computer, and Azure IoT.
• Updating the device code sample to include our connection data and send it to Azure IoT to be viewed remotely.

• An SSH client – This makes it so you can remotely access the Raspberry Pi’s command line remotely from your computer
  • Windows doesn’t have a built-in SSH client. We recommend using PuTTY
  • Many Linux distributions and Mac OS has SSH built into their terminal. If yours does not, we recommend OpenSSH
  • See also: SSH Using Linux or Mac OS

• Microsoft Azure IoT Starter Kit
  • 8GB MicroSD Card (Comes with the kit, flashed with Windows 10 IoT Core)
  • A USB Mini cable
  • An Ethernet cable or Wi-Fi dongle

• Log in to Azure IoT Suite with your Microsoft account and click Create a New Preconfigured Solution

Note: Creating an account is free. Click here to get your Azure free trial

• Click select in the Remote Monitoring option
• Type in a solution name. For this tutorial we’ll be using “Pi2Suite”

Note: Make sure to copy down the names and connection strings mentioned into a text document for reference later.

• Choose your subscription plan and geographic region, then click Create Solution
• Wait for Azure to finish provisioning your IoT suite (this process may take up to 10 minutes), and then click Launch

Note: You may be asked to log back in. This is to ensure your solution has proper permissions associated with your account.

• Open the link to your IoT Suite’s “Solution Dashboard.” You may have been redirected there already.
• This opens your personal remote monitoring site at the URL <Your IoT Hub suite name>.azurewebsites.net (e.g. Pi2Suite.azurewebsites.net)
• Click Add a Device at the lower left hand corner of your screen
• Add a new custom device
• Enter your desired device ID. In this case we’ll use “RaspPi”, and then click Create
• Make note of your device ID, Device Key, and IoT Hub Hostname to enter into the code you’ll run on your device later

Warning: The Remote Monitoring solution provisions a set of Azure IoT Services in your Azure account. To avoid unnecessary Azure consumption, you can do two things:

1. Disable the simulated devices created with the solution (Go to Devices>>Select the device>> on the device details menu on the right, clich on Disable Device. Repeat with all the simulated devices).

2. Stop your remote monitoring solution while you are working on the next steps. (See: Troubleshooting)

##1.4 Prepare the Device

If this is the first time you are using Raspberry Pi 2, now is the time to set it up. If you’ll be using Windows, please use windowsondevices.com for detailed guidelines on how to get started with the Raspberry Pi. If you’re using Linux, Raspberry Pi and Adafruit have a set of tutorials and videos to help you get started.

Please visit the following links:

• Raspberry Pi NOOBS setup page
• Using this image as a reference, connect your BME280 and Feather M0 WiFi to the breadboard

Note: Column on the left corresponds to the sensor and on the right to the board. On the image, the board is place between 10 and 30 and sensor between 1 and 9.

At the end of your work, your Raspberry Pi should be connected with a working sensor.

The default login for Raspbian is username pi with password raspberry. If you use the Raspbian interface directly, in the task bar up top, you can launch the Terminal using the 3rd icon from the left – The one that looks like a monitor.

• You need to discover the IP address of your Raspberry Pi before you can connect using PuTTY. Type hostname -I in a command prompt to discover your IP. For more information see: https://www.raspberrypi.org/documentation/remote-access/ip-address.md
• With the Raspberry Pi on and running, open an SSH terminal program such as PuTTY on your desktop machine.
• Use the IP address from previous step as the Host Name, Port=22, and Connection type=SSH to complete the connection.
• When prompted, log in with username pi, and password raspberry.

• See: SSH Using Linux of Mac OS

Now we can download the Remote Monitoring device sample, input your device credentials (i.e. your "Connection String") and start sending data from the Pi to our Azure Remote Monitoring solution.

Clone the following repositories and combine them by entering the following commands on your Pi:

cd ~
git clone --recursive https://github.com/Azure/azure-iot-sdks.git
git clone https://github.com/Azure-Samples/iot-hub-c-raspberrypi-getstartedkit.git
cp -a iot-hub-c-raspberrypi-getstartedkit/azure-iot-sdks/. azure-iot-sdks/

Now, we need to update device connection string

• Edit the file by entering the following command:

nano ~/azure-iot-sdks/c/serializer/samples/remote_monitoring/remote_monitoring.c

• Use the arrow keys to navigate to the following lines:

static const char* deviceId = "[Device Id]";
static const char* deviceKey = "[Device Key]";
static const char* hubName = "[IoTHub Name]";
static const char* hubSuffix = "[IoTHub Suffix, i.e. azure-devices.net]";

• Replace the placeholder with your device and IoT Hub information you created and saved at the beginning of this tutorial.
• Save and exit with Control-o, Enter, Control-x

Now that you’re exited out of Nano, install the prerequisite packages for the Microsoft Azure IoT Device SDK for C by issuing the following commands using the terminal (either on the device or through an SSH client like PuTTY):

sudo apt-get update
sudo apt-get install curl libcurl4-openssl-dev uuid-dev uuid g++ make cmake git unzip openjdk-7-jre libssl-dev libncurses-dev subversion gawk

• Now go ahead and build the updated sample solution by entering the following:

cd ~/azure-iot-sdks/
sudo ~/azure-iot-sdks/c/build_all/linux/setup.sh
chmod +x ~/azure-iot-sdks/c/build_all/linux/build.sh
~/azure-iot-sdks/c/build_all/linux/build.sh

• Before running the program we need to enable the SPI drive to install by default at boot.

sudo nano /boot/config.txt

• Scroll down and find the line:

#dtparam=spi=on

• Delete the # at the beginning of the line to uncomment it.
• Save and exit with Control-o, Enter, Control-x, then run the command:

sudo reboot

• Now that everything is compiled, it’s time to run the program. Enter the command:

sudo ~/cmake/serializer/samples/remote_monitoring/remote_monitoring

If all goes well, you will begin to see data streaming! Press ctrl-c to exit at any time. Below is a sample of the expected command prompt output:

IoTHubClient accepted the message for delivery
Humidity = 48.4% Temperature = 23.9*C
Sending sensor value Temperature = 2, Humidity = 1610612736
IoTHubClient accepted the message for delivery

• Once you have the sample running, visit your dashboard by visiting azureiotsuite.com and clicking “Launch” on your solution
• Make sure the “Device to View” in the upper right is set to your device
• If the demo is running, you should see the graph change as your data updates in real time!

Note: Make sure to stop your remote monitoring solution once you have completed this to avoid unnecessary Azure consumption! (See: Troubleshooting)

Please visit our Azure IoT Dev Center for more samples and documentation on Azure IoT.

This tutorial describes the process of taking your Microsoft Azure IoT Starter Kit for Raspberry Pi 2, and using it to develop a temperature and humidity reader that can communicate with Microsoft Azure IoT Services, process the data, detects abnormal data, and sends that back to the Pi for use. We will be using the Raspbian OS and Microsoft Azure IoT SDK. For Windows 10 IoT Core samples, please visit windowsondevices.com.

• 2.1 Tutorial Overview
• 2.2 Requirements
  • 2.2.1 Required Software
  • 2.2.2 Required Hardware
• 2.3 Create a New Microsoft Azure IoT Hub and Add Device
• 2.4 Prepare the Device
  • 2.4.1 Log in and Access the Terminal
  • 2.4.2 Log in Using PuTTY
• 2.5 Create an Event Hub
• 2.6 Create a Storage Account for Table Storage
• 2.7 Create a Stream Analytics job to Save Sensor Data in Table Storage and Raise Alerts
• 2.8 Node Application Setup
• 2.9 Configure the Command Center device sample
  • 2.9.1 Update the Connection Data
• 2.10 Build the Modified Sample
• 2.11 Next steps

This tutorial has the following steps:

• Provision an IoT Hub resource on Microsoft Azure and adding your device.
• Prepare the device, get connected to the device, and set it up so that it can read sensor data.
• Configure your Microsoft Azure IoT services by adding Event Hub, Storage Account, and Stream Analytics resources.
• Prepare your local web solution for monitoring and sending commands to your device.
• Update the sample code to respond to commands and include the data from our sensors, sending it to Microsoft Azure to be viewed remotely.

The end result will be a functional command center where you can view the history of your device's sensor data, a history of alerts, and send commands back to the device.

• Git - For cloning the required repositories
• Node.js - For the Node application, we will go over this later.
• An SSH client – This makes it so you can remotely access the Raspberry Pi’s command line remotely from your computer
  • Windows doesn’t have a built-in SSH client. We recommend using PuTTY
  • Many Linux distributions and Mac OS has SSH built into their terminal. If yours does not, we recommend OpenSSH
  • See also: SSH Using Linux on Mac OS

• Microsoft Azure IoT Starter Kit
  • 8GB MicroSD Card (comes with the kit, flashed with Windows 10 IoT Core)
  • A USB Mini cable
  • An Ethernet cable or Wi-Fi dongle

• To create your Microsoft Azure IoT Hub and add a device, follow the instructions outlined in the Setup IoT Hub Microsoft Azure Iot SDK page.
• We named our IoT Hub raspPiIoT

Note: Make sure to copy down the names and connection strings mentioned into a text document for reference later.

• After creating your device, make note of your connection string to enter into the code you’ll run on your device later

If this is the first time you are using Raspberry Pi 2, now it’s the time to set it up. If you’ll be using Windows, please use windowsondevices.com for detail guidelines on how to get started with the Raspberry Pi. If you’re using Linux, Raspberry Pi and Adafruit have a set of tutorials and videos to help you get started.

Please visit the following links:

• Raspberry Pi NOOBS setup page
• Using this image as a reference, connect your BME280 to the breadboard.

Note: Column on the left corresponds to the sensor and on the right to the board. On the image, the board is place between 10 and 30 and sensor between 1 and 9. The - symbol refers to the blue row on the board, with the pins counting from the left starting at 1. See the diagram above for more reference.

At the end of your work, your Raspberry Pi should be connected with a working sensor.

The default login for Raspbian is username pi with password raspberry. If you use the Raspbian interface directly, in the task bar up top, you can launch the Terminal using the 3rd icon from the left – The one that looks like a monitor.

• You need to discover the IP address of your Raspberry Pi before you can connect using PuTTY. Type hostname -I in a command prompt to discover your IP.For more information see: https://www.raspberrypi.org/documentation/remote-access/ip-address.md
• With the Raspberry Pi on and running, open an SSH terminal program such as PuTTY on your desktop machine.
• Use the IP address from previous step as the Host Name, Port=22, and Connection type=SSH to complete the connection.
• When prompted, log in with username pi, and password raspberry.

Before creating the Event Hub, we need to have a Resource Group created that will contain our resources. In the Microsoft Azure Portal:

• Click the "Resource groups" button on the left
• Click the "+" icon (Add)
  • Enter the name for your resource group (We chose Pi2Suite)
  • Subscription: Your choice
  • Region: Your choice

Event Hub is an Azure IoT publish-subscribe service that can ingest millions of events per second and stream them into multiple applications, services or devices.

• From the Azure Portal look at the top left portion of the screen
• Click on New -> Internet of Things-> Event Hub
• After being redirected, click "Custom Create", Enter the following settings for the Event Hub (use a name of your choice for the event hub and the namespace):
  • Event Hub Name: piEventHub
  • Region: Your choice
  • Subscription: Your choice
  • Namespace Name: Your Project Namespace, in our case “Pi2Suite”
• Click the arrow to continue.
• Choose to create 4 partitions and retain messages for 7 days.
• Click the check at the bottom right hand corner to create your event hub.
• Click on your Pi2Suite service bus (what you named your service bus)
• Click on the Event Hubs tab
• Select the piEventHub Event Hub and go in the Configure tab in the Shared Access Policies section, add a new policy:
  • Name = readwrite
  • Permissions = Send, Listen
• Click Save at the bottom of the page, then click the Dashboard tab near the top and click on Connection Information at the bottom
• Copy down the connection string for the readwrite policy you created.
• From the your IoT Hub Settings (The Resource that has connected dots) on the Microsoft Azure Portal, click the Messaging blade (found in your settings), write down the Event Hub-compatible name
• Look at the Event-hub-compatible Endpoint, and write down this part: sb://thispart.servicebus.windows.net/ we will call this one the IoTHub EventHub-compatible namespace

Now we will create a service to store our data in the cloud.

• Log on to the Microsoft Azure Portal
• In the menu, click New and select Data + Storage then Storage Account
• Choose Classic for the deployment model and click on Create
• Enter the name of your choice (We chose pistorage) for the account name, Standard-RAGRS for the type, choose your subscription, select the resource group you created earlier, then click on Create
• Once the account is created, find it in the resources blade or click on the pinned tile, go to Settings, Keys, and write down the primary connection string.

Stream Analytics is an Azure IoT service that streams and analyzes data in the cloud. We'll use it to process data coming from your device.

• Log on to the Microsoft Azure Portal

• In the menu, click New, then click Internet of Things, and then click Stream Analytics Job

• Enter a name for the job (We chose “PiStorageJob”), a preferred region, then choose your subscription. At this stage you are also offered to create a new resource group or to use an existing resource group. Choose the resource group you created earlier (In our case, Pi2Suite).

• Once the job is created, open your Job’s blade or click on the pinned tile, and find the section titled “Job Topology” and click the Inputs tile. In the Inputs blade, click on Add

• Enter the following settings:

  • Input Alias = TempSensors
  • Source Type = Data Stream
  • Source = IoT Hub
  • IoT Hub = raspPiIoT (use the name for the IoT Hub you create before)
  • Shared Access Policy Name = iothubowner
  • Shared Access Policy Key = The iothubowner primary key can be found in your IoT Hub Settings -> Shared access policies
  • IoT Hub Consumer Group = "" (leave it to the default empty value)
  • Event serialization format = JSON
  • Encoding = UTF-8

• Back to the Stream Analytics Job blade, click on the Query tile (next to the Inputs tile). In the Query settings blade, type in the below query and click Save:

SELECT
    DeviceId,
    EventTime,
    MTemperature as TemperatureReading
INTO
    TemperatureTableStorage
from TempSensors
WHERE
   DeviceId is not null
   and EventTime is not null

SELECT
    DeviceId,
    EventTime,
    MTemperature as TemperatureReading
INTO   
    TemperatureAlertToEventHub
FROM
    TempSensors
WHERE MTemperature>25 

Note: You can change the 25 to 0 when you're ready to generate alerts to look at. This number represents the temperature in degrees Celsius to check for when creating alerts. 25 degrees Celsius is 77 degrees Fahrenheit.

• Back to the Stream Analytics Job blade, click on the Outputs tile and in the Outputs blade, click on Add
• Enter the following settings then click on Create:
  • Output Alias = TemperatureTableStorage
  • Sink = Table Storage
  • Storage account = pistorage (The storage you made earlier)
  • Storage account key = (The primary key for the storage account made earlier, can be found in Settings -> Keys -> Primary Access Key)
  • Table Name = TemperatureRecords*Your choice - If the table doesn’t already exist, Local Storage will create it
  • Partition Key = DeviceId
  • Row Key = EventTime
  • Batch size = 1
• Back to the Stream Analytics Job blade, click on the Outputs tile, and in the Outputs blade, click on Add
• Enter the following settings then click on Create:
  • Output Alias = TemperatureAlertToEventHub
  • Source = Event Hub
  • Service Bus Namespace = Pi2Suite
  • Event Hub Name = pieventhub (The Event Hub you made earlier)
  • Event Hub Policy Name = readwrite
  • Event Hub Policy Key = Primary Key for readwrite Policy name (That's the one you wrote down after creating the event hub)
  • Partition Key Column = 0
  • Event Serialization Format = JSON
  • Encoding = UTF-8
  • Format = Line separated
• Back in the** Stream Analytics blade**, start the job by clicking on the **Start **button at the top

Note: Make sure to stop your Command Center jobs once you have when you take a break or finish to avoid unnecessary Azure consumption! (See: Troubleshooting)

• If you do not have it already, install Node.js and NPM.
  • Windows and Mac installers can be found here: https://nodejs.org/en/download/
    • Ensure that you select the options to install NPM and add to your PATH.
  • Linux users can use the commands:

sudo apt-get update
sudo apt-get install nodejs
sudo apt-get install npm

• Additionally, make sure you have cloned the project repository locally by issuing the following command in your desired directory:

git clone https://github.com/Azure-Samples/iot-hub-c-raspberrypi-getstartedkit.git

• Open the command_center_node folder in your command prompt (cd path/to/command_center_node) and install the required modules by running the following:

npm install -g bower
npm install 
bower install

• Open the config.json file and replace the information with your project. See the following for instructions on how to retrieve those values.

  • eventhubName:
    • Open the Classic Azure Management Portal
    • Open the Service Bus namespace you created earlier
    • Switch to the EVENT HUBS page
    • You can see and copy the name of your event hub from that page
  • ehConnString:
    • Click on the name of the event hub from above to open it
    • Click on the "CONNECTION INFORMATION" button along the bottom.
    • From there, click the button to copy the readwrite shared access policy connection string.
  • deviceId:
    • Use the information on the Manage IoT Hub to retrieve your deviceId using either the Device Explorer or iothub-explorer tools.
  • iotHubConnString:
    • In the Azure Portal
    • Open the IoT Hub you created previously.
    • Open the "Settings" blade
    • Click on the "Shared access policies" setting
    • Click on the "service" policy
    • Copy the primary connection string for the policy
  • storageAccountName:
    • In the Azure Portal
    • Open the classic Storage Account you created previously to copy its name
  • storageAccountKey:
    • Click on the name of the storage account above to open it
    • Click the "Settings" button to open the Settings blade
    • Click on the "Keys" setting
    • Click the button next to the "PRIMARY ACCESS KEY" top copy it
  • storageTableName:
    • This must match the name of the table that was used in the Stream Analytics table storage output above.
    • If you used the instructions above, you would have named it TemperatureRecords
    • If you named it something else, enter the name you used instead.

{
    "port": "3000",
    "eventHubName": "event-hub-name",
    "ehConnString": "Endpoint=sb://name.servicebus.windows.net/;SharedAccessKeyName=readwrite;SharedAccessKey=aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa=",
    "deviceId": "iot-hub-device-name",
    "iotHubConnString": "HostName=iot-hub-name.azure-devices.net;SharedAccessKeyName=service;SharedAccessKey=aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa=",
    "storageAcountName": "aaaaaaaaaaa",
    "storageAccountKey": "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa==",
    "storageTable": "TemperatureRecords"
} 

• Now it is time to run it! Enter the following command:

node server.js

• You should then see something similar to:

app running on http://localhost:3000
client connected

• Visit the url in your browser and you will see the Node app running!

To deploy this project to the cloud using Azure, you can reference Creating a Node.js web app in Azure App Service.

Next, we will update your device so that it can interact with all the things you just created.

Now we can download the sample solution repos and incorporate the data from the sensors and send it up to our Microsoft Azure Command Center solution.

Clone the repositories by entering the following commands on your Pi:

cd ~
git clone --recursive https://github.com/Azure/azure-iot-sdks.git
git clone https://github.com/Azure-Samples/iot-hub-c-raspberrypi-getstartedkit.git
cp -a iot-hub-c-raspberrypi-getstartedkit/azure-iot-sdks/. azure-iot-sdks/

Now, we need to update three things:

• Include the required libraries (included in the git repository)
• Include the Device ID
• Update the connection string to use the string we saved above

To update the connection data, we need to update the source file. You can edit the file by entering the following command:

nano ~/azure-iot-sdks/c/serializer/samples/simplesample_amqp/simplesample_amqp.c

• Use the arrow keys to navigate to the following lines:

static const char* connectionString = "HostName=<host_name>.azure-devices.net;DeviceId=<device_id>;SharedAccessKey=<device_key>";

• Replace the placeholder with your device connection information you gathered at the beginning of this tutorial.
• Save and exit with Control-o, Enter, Control-x

Now that you’re exited out of Nano, install the prerequisite packages for the Microsoft Azure IoT Device SDK for C by issuing the following commands using the terminal (either on the device or through an SSH client like PuTTY):

sudo apt-get update
sudo apt-get install curl libcurl4-openssl-dev uuid-dev uuid g++ make cmake git unzip openjdk-7-jre libssl-dev libncurses-dev subversion gawk

• Now go ahead and build the updated sample solution by entering the following:

cd ~/azure-iot-sdks/
sudo ~/azure-iot-sdks/c/build_all/linux/setup.sh
chmod +x ~/azure-iot-sdks/c/build_all/linux/build.sh
~/azure-iot-sdks/c/build_all/linux/build.sh

• Now that everything is compiled, it’s time to run the program. Enter the command:

sudo cmake/serializer/samples/simplesample_amqp/simplesample_amqp

You will now see data being sent off at regular intervals to Microsoft Azure. An alert has also been set to go off when it detects the temperature is above 25 degrees celsius (77'F). You can cup your hand around the sensor and blow warm air to raise the temperature and when the alert goes off, you will see the LED you’ve set up turn red!

Note: If you're in a particularly hot or cold room, you may need to adjust the alert temperature specified in Create a Stream Analytics job to Save Sensor Data in Table Storage and Raise Alerts.

Head back to your Node application and you will have a fully functional command center, complete with a history of sensor data, alerts that display when the temperature got outside a certain range, and commands that you can send to your device remotely.

Note: Make sure to stop your Command Center jobs once you have when you finish to avoid unnecessary Azure consumption! (See: Troubleshooting)

Please visit our Azure IoT Dev Center for more samples and documentation on Azure IoT.

• In the Microsoft Azure Portal
  • Click on "All Resources"
  • For each Stream Analytics and Web App resource:
    • Click on the resource and click the "Stop" button in the new blade that appears
  • For each IoT Hub resource:
    • Click on the resource and click the "Devices" button in the new blade that appears
    • Click on each device in the list and click the "Disable" button that appears in the new blade at the bottom