Build a containerized Python web app in Azure

Create an Azure Container Registry

If you have an existing Azure Container Registry you wish to use, skip this next step and proceed to the next step. Otherwise, create a new Azure Container Registry using the Azure CLI.

Azure CLI commands can be run in the Azure Cloud Shell or in your local development environment with the Azure CLI installed.

1. Create an Azure container registry with the az acr create command.

   • Bash
   • PowerShell

   #!/bin/bash
   # Use the resource group that you created in part 2 of this tutorial series.
   RESOURCE_GROUP_NAME='msdocs-web-app-rg'
   # REGISTRY_NAME must be unique within Azure and contain 5-50 alphanumeric characters.
   REGISTRY_NAME='msdocscontainerregistryname'
   
   echo "Creating Azure Container Registry $REGISTRY_NAME..."
   az acr create -g $RESOURCE_GROUP_NAME -n $REGISTRY_NAME --sku Standard
   

   In the JSON output of the command, locate the loginServer value. This value represents the fully qualified registry name (all lowercase) and contains the registry name.

2. If you're using the Azure CLI on your local machine, execute the az acr login command to log in to the container registry.

   az acr login -n $REGISTRY_NAME
   

   The command adds "azurecr.io" to the name to create the fully qualified registry name. If successful, you see the message "Login Succeeded".

   Note

   In the Azure Cloud Shell, the az acr login command isn't necessary, as authentication is handled automatically through your Cloud Shell session. However, if you encounter authentication issues, you can still use it.

Build an image in Azure Container Registry

You can generate the container image directly in Azure through various approaches:

• The Azure Cloud Shell allows you to construct the image entirely in the cloud, independent of your local environment.
• Alternatively, you can use VS Code or the Azure CLI to create it in Azure from your local setup, without needing Docker to be running locally.

Azure CLI commands can be run in your local development environment with the Azure CLI installed or in Azure Cloud Shell.

1. In the console, navigate to the root folder for your cloned repository from part 2 of this tutorial series.

2. Build the container image using the az acr build command.

   az acr build -r $REGISTRY_NAME -g $RESOURCE_GROUP_NAME -t msdocspythoncontainerwebapp:latest .
   # When using Azure Cloud Shell, run one of the following commands instead:
   # az acr build -r $REGISTRY_NAME -g $RESOURCE_GROUP_NAME -t msdocspythoncontainerwebapp:latest https://github.com/Azure-Samples/msdocs-python-django-container-web-app.git
   # az acr build -r $REGISTRY_NAME -g $RESOURCE_GROUP_NAME -t msdocspythoncontainerwebapp:latest https://github.com/Azure-Samples/msdocs-python-flask-container-web-app.git
   

   The last argument in the command is the fully qualified path to the repo. When running in Azure Cloud Shell, use https://github.com/Azure-Samples/msdocs-python-django-container-web-app.git for the Django sample app and https://github.com/Azure-Samples/msdocs-python-flask-container-web-app.git for the Flask sample app.

3. Confirm the container image was created with the az acr repository list command.

   az acr repository list -n $REGISTRY_NAME
   

Create an Azure Container Registry

If you have an existing Azure Container Registry you wish to use, skip this next step and proceed to the next step. Otherwise, create a new Azure Container Registry using VS Code.

1. In the Docker extension in VS Code, go to REGISTRIES and click Azure to connect to the Azure Container Registry.

2. In Visual Studio Code, select F1 or CTRL+SHIFT+P to open the command palette. Then type "registry" and select the Azure Container Registry: Create Registry task.

   Alternatively, in the Docker extension REGISTRIES section, right-click your subscription, and select Create Registry. This action starts the same create registry task.

3. Follow the prompts and enter the following values:

   • Registry name: Enter msdocscontainerregistryname. The registry name must be unique within Azure, and contain 5-50 alphanumeric characters.

   • SKU: Select Standard.

   • Resource group: Use your existing resource group.

4. In the Docker extension, in the REGISTRIES section, find the registry you created, right-click, and select View Properties.

   Look for the loginServer key value pair in the output. The value is the fully qualified name of the registry.

Build an image in Azure Container Registry

You can generate the container image directly in Azure through various approaches:

• The Azure Cloud Shell allows you to construct the image entirely in the cloud, independent of your local environment.
• Alternatively, you can use VS Code or the Azure CLI to create it in Azure from your local setup, without needing Docker to be running locally.

These steps require that VS Code is opened in the working folder of your web app.

1. In the Docker extension, click REGISTRIES and then click Azure to connect to the Azure Container Registry.

   

2. Select F1 or CTRL+SHIFT+P to open the command palette, type "registry", and select the Azure Container Registry: Build Image in Azure task.

   If you don't see the task, make sure that Azure appears under REGISTRIES in the Docker extension. You can also right-click the Dockerfile and select Build Image in Azure to run the task.

3. Input the following values and follow the prompts to create the image.

   • Tag image: Use the image name "msdocspythoncontainerwebapp:latest".
   • Registry provider: Select Azure.
   • Subscription: Select the subscription you used to create the Azure Container Registry.
   • Registry: Select the registry you created or one to which you have access.
   • Base OS image: Select Linux.

   Check the OUTPUT window for progress and information on the build. If you get a credentials error, right-click the registry in the REGISTRIES section of the Docker extension and select Refresh.

4. Confirm the image in the Azure Container Registry.

   1. In the Docker extension, in the REGISTRIES section, find the container image created. You may need to close and re-open VS Code to see the image.

   2. Confirm the name and tag latest.