Skip to content

Latest commit

 

History

History
147 lines (103 loc) · 5.47 KB

README.md

File metadata and controls

147 lines (103 loc) · 5.47 KB
Raw

Azure IoT Edge Project

This project is for the development of IoT Edge modules.

Start Here

For anyone new to IoT Edge, it is recommended you follow this tutorial first to get everything setup, and to gain a better understanding of how IoT Edge works.

Folder Contents

Here is a general guide of the project's folder structure for reference.

  • .vscode | Folder containing launch settings for running the project in VSCode
  • modules | Folder containing all Edge module code
    • individual module | Each module should have its own folder containing
      • requirements.txt
      • module.json
      • Dockerfile
      • main.py
  • tests - folder for Edge module unit and integration tests
    • filename guidance test_<moduleName>

Deployment Manifest Templates

Deployment manifest templates contain the information about which modules should be deployed into an IoT Edge device, and their configurations. Only one manifest can be deployed on a device at any one time. Deploying a manifest will overwrite any other modules running on the device.

Setup

Environment Variables

Create a .env file containing:

SUBSCRIPTION_ID="<azure_subscription_id>"
AAD_TENANT_ID="<tenant_id>"
AAD_SERVICE_PRINCIPAL_ID="<service_principal_id>"
AAD_SERVICE_PRINCIPAL_SECRET="<service_principal_secret>"

RESOURCE_GROUP="<resource_group>"
AMS_ACCOUNT="<media_service_account>"
APPINSIGHTS_INSTRUMENTATIONKEY="<app insights instrumentation key>"

ACR_USER="<acr_username>"
ACR_PASSWORD="<acr_password>"
ACR_ADDRESS="<acr_username>.azurecr.io"
IMAGE_TAG="<docker_image_tag>"

# Container insights settings (follow documentation here for initial setup: https://labs.iotedge.dev/codelabs/edgemon-preview/#0)
IOT_HUB_RESOURCE_ID="/subscriptions/<SUBSCRIPTION_ID>/resourceGroups/<RESOURCE_GROUP>/providers/Microsoft.Devices/IotHubs/<IOTHUB_NAME>"
LOG_ANALYTICS_WORKSPACE_ID="<Log Analytics Workspace ID>"
LOG_ANALYTICS_WORKSPACE_KEY="<Log Analytics Workspace key>"
CONTAINER_INSIGHTS_ENDPOINTS="http://edgeHub:9600/metrics,http://edgeAgent:9600/metrics,http://lvaEdge:9600/metric"

VIDEO_INPUT_FOLDER_ON_DEVICE="/home/<YOUR_ADMIN_USERNAME>/samples/input"
VIDEO_OUTPUT_FOLDER_ON_DEVICE="/home/<YOUR_ADMIN_USERNAME>/samples/output"

In our DevOps pipelines <YOUR_ADMIN_USERNAME> is usually set to edgeadmin

Generate manifest

To generate a deployment manifest from the template, right click on the template file and select Generate IoT Edge deployment manifest. This will create the corresponding deployment manifest file in ./config folder.

Make sure you have the Azure IoT Tools extension installed

Deploy

Once you have the manifest in the ./config folder, right click on the manifest file and select Create Deployment for Single Device.

Debug

  • Open this project using VSCode, specifically from the edge/ folder

  • Make sure port 5678 is open on your edge device

  • Add this to the import section of the edge module main.py file

    import ptvsd
ptvsd.enable_attach(('0.0.0.0',  5678))

  • Replace 0.0.0.0 with the public IP address of the edge device

  • Add breakpoint() to the function you want to break on when debugging, for example

    async def message_handler(message):
  breakpoint() # I am adding a breakpoint here
  if message.input_name == "detectedObjects":
      logging.debug("Message received on detectedObjects route")

  • In the .vscode/launch.json file in this folder, update the host value with the public IP address of the edge device

  • Right-click on the deployment.debug.template.json and select Build and Push IoT Edge solution

  • In the config/ folder, right-click on the deployment.debug.amd64.json file and select Create Deployment for a Single Device, then select your edge device

  • Navigate to the debug tab in VS Code and run the debugger ObjectDetectionBusinessLogic Module Remote Debug (Python)

Log Level

You can set the log level of the business logic module using the logLevel module property.

  "objectDetectionBusinessLogic": {
    "properties.desired": {
      "logLevel": "INFO"
    }
  }

Accepted values:

  • DEBUG
  • INFO
  • WARNING
  • ERROR
  • CRITICAL

Troubleshooting the LVA Edge Module

There is a great troubleshooting guide provided by the LVA product team, you can read the documentation here.

One thing you might want to consider is adding MediaPipeline log category in the deployment manifest, you can update the logCategories property under the lvaEdge desired properties section of the manifest. This log category is already enabled as part of our debug deployment manifest. Read more about LVA logging here.

MediaPipeline: Low-level logs that might offer insight when you're troubleshooting problems, like difficulties establishing a connection with an RTSP-capable camera.