Azure IoT PCS CLI Overview

Command Line Interface for deploying an IoT pre-configured solution into a user's subscription.

This CLI has the ability to deploy two configurations of PCS solutions:

1. Basic - deploys all resources to a single VM.
2. Enterprise - deploys resources using Azure Container Service and Kubernetes across multiple VMs.

Dependencies

The command line interface depends on:

• nodejs used as the runtime for the CLI. Please install node before attempting a deployment.

Basic

The purpose of the basic deployment is to demo the capabilities of the system and requires minimal setup, deploying all resources to a single VM.

Basic deployment provisions following resources:

1. Azure IoT Hub
2. Azure Cosmos DB
3. Azure Storage
4. Single instance of Azure Virtual Machine with Docker Extension

At the end of deployment, Remote Monitoring WebApp and all the microservices are ready for demo pursposes.

Enterprise

The Enterprise deployment offers a production ready deployment that can be scaled up or down as needed. It uses Azure Container Service and Kubernetes for orchestration. It would be nice to have installed kubectl for running commands on kubernetes in addition to pcs.

Enterprise deployment provisions following resources:

1. Azure IoT Hub
2. Azure Cosmos DB
3. Azure Container Service which also provisions following:
   1. Azure Storage
   2. Three instances of Azure Virtual Machine with Docker

How to use the CLI

Install CLI

npm install iotsuite-cli -g

Basic Deployment

Deploy Azure Resources

1. If you haven't logged in with your Azure account from the command prompt run pcs login.
2. Run either pcs or pcs -t remotemonitoring -s basic. These are equivalent in that they will both deploy a basic deployment (i.e. a deployment to a single VM).
3. Follow the on-screen prompts
4. The results of the deployment will be saved to a file named output.json

Verify the Web UI and Microservices are deployed

Click on the link that is shown in the output window, it will take you to the Remote Monitoring WebApp

Enterprise Deployment

Deploy Azure Resources

1. pcs -t remotemonitoring -s standard
2. Follow the on-screen prompts
3. The results of the deployment will be saved to a file named {deployment-name}-output.json

Sample output format:

"resourceGroup" : {
    "type": "string",
    "value": "{myResourceGroupName}"
},
"iotHubHostName": {
    "type": "string",
    "value": "{myIoTHubHostName}"
},
"iotHubConnectionString": {
    "type": "string",
    "value": "{HostName={hubname}.azure-devices.net;
    SharedAccessKeyName={policy type};SharedAccessKey={Access Key};}"
},
"documentDBConnectionString" : {
    "type": "string",
    "value": "{AccountEndpoint={URI};AccountKey={Key};}"
}

Verify the Web UI and Microservices are deployed

1. Click on the link that is shown in the output window, it will take you to the Remote Monitoring WebApp
2. It can take upto 5 minutes for the webapp to be ready
3. Go to {azurewebitesurl}/hubmanager/v1/status to see HubManager microservice status
4. Go to {azurewebitesurl}/devices/v1/status to see Devices microservice status

Configuration

Kubernetes Dashboard

To view Kubernetes dashboard, run the following command, which will start a local web proxy for your cluster (it will start a local server at http://127.0.0.1:8001/ui):

kubectl proxy

CLI Options

To get help run pcs -h or --help
To get the version run pcs -v or --version

Feedback

Please enter issues, bugs, or suggestions as GitHub Issues here: https://github.com/Azure/pcs-cli/issues.

Related

• Contributing and Development setup