Challenge-06.md

Challenge 6 - Dapr MQTT Input Binding

< Previous Challenge - Home - Next Challenge >

Introduction

In this challenge, you're going to add a Dapr input binding in the TrafficControlService. It'll receive entry- and exit-cam messages over the MQTT protocol.

Description

In this challenge you'll focus on Dapr input bindings. The following diagram depicts how input bindings work:

For this hands-on challenge, you will add an input binding leveraging the Dapr binding building block. In the previous challenge, you implemented a Dapr output binding.

• Stand up a Mosquitto MQTT message broker in a Docker container.
• Modify the TrafficControlService (TrafficController class) to use the Dapr MQTT input binding to receive entry-cam and exit-cam messages over the MQTT protocol.
• Modify the Simulation app to put trafficcontrol/entrycam & trafficcontrol/exitcam messages on the MQTT queue.
  • Create a new class called MqttTrafficControlService to do this (look at the HttpTrafficControlService as an example).
  • Modify the Program class to use this new service.
• Create a Dapr configuration file for specifying the Dapr SMTP input binding components.
• Restart all services & run the Simulation application.
• Once you have the solution running locally, modify the code to use Azure IoTHub & EventHub as the MQTT message broker.

Success Criteria

This challenge targets the operation labeled as number 5 in the end-state setup:

Local

Azure

• Validate that the Mosquitto MQTT message broker service is running locally.
• Validate that the TrafficControlService receives messages via its Dapr component.
• Validate that the Simulation application publishes entry-cam and exit-cam messages to the MQTT broker.
• Validate that messages are being sent through the Azure IoT Hub & EventHub.

DIY instructions

• Use MQTT broker Mosquitto for local development.

• In order to connect to Mosquitto, you need to pass in a custom configuration file when starting it. With Docker, you can pass a configuration file when starting a container using a Volume mount. The folder Resources/Infrastructure/mosquitto already contains a config file you can use.

  1. Open a terminal window in VS Code and make sure the current folder is Resources/Infrastructure/mosquitto.

  2. Start a Mosquitto MQTT container by entering the following command: When running on Windows:

     docker run -d -p 1883:1883 -p 9001:9001 -v $pwd/:/mosquitto/config/ --name dtc-mosquitto eclipse-mosquitto

     When running on Mac or Linux:

     docker run -d -p 1883:1883 -p 9001:9001 -v $(pwd)/:/mosquitto/config/ --name dtc-mosquitto eclipse-mosquitto

  This will pull the docker image eclipse-mosquitto from Docker Hub and start it. The name of the container will be dtc-mosquitto. The server will be listening for connections on port 1883 for MQTT traffic.

  The -v flag specifies a Docker volume mount. It mounts the current folder (containing the config file) as the /mosquitto/config/ folder in the container. Mosquitto reads its config file from that folder.

• To peak into the Mosquitto server, open a new terminal window and execute the following command:

  docker logs dtc-mosquitto

• Add the following MQTT configuration flags when connecting to the Mosquitto MQTT queue.

  var configuration = new MqttConfiguration()
  {
    KeepAliveSecs = 60,
    Port = 1883
  };

• Use Azure IoT Hub & EventHub for deployments to Azure.

  • Create a IoT Device in Azure IoT Hub to represent your Simulation app.

    az iot hub device-identity create --device-id simulation --hub-name <iot-hub-name>

  • Get the IoT Hub Connection String for the device you just created.

    az iot hub device-identity connection-string show --device-id simulation --hub-name <iot-hub-name>

  • Use the Microsoft.Azure.Devices.Client NuGet package to connect to the IoT Hub instead of the local MQTT broker.

Learning Resources

• Introduction to this building block
• Bindings chapter
• Dapr for .NET developers
• Azure IoT Hub