Creating an Azure Compute Gallery and Capturing the Image of a Virtual Machine (VM)

Azure Compute Gallery is a powerful feature in Azure that enables users to efficiently manage and share VM images across multiple subscriptions and regions. It provides a centralized location to store images and simplifies the deployment of consistent environments across your organization or across your Azure infrastructure. In this article, we will walk through the steps of creating an Azure Compute Gallery and capturing the image of a VM, with an emphasis on the state of the VM being specialized. We’ll also discuss when you might want to generalize the VM instead.

Prerequisites:

Before beginning, make sure you have the following:

• An active Azure subscription.
• Azure CLI or Azure PowerShell installed and configured.
• Basic understanding of Azure Virtual Machines and image creation.
• A VM in Azure that is in the desired configuration for capturing.

---

Step 1: Create an Azure Compute Gallery

First, let’s set up the Azure Compute Gallery to store and manage your VM images.

1. Navigate to Azure Portal: Go to the Azure Portal and log into your account.

1. Create a Compute Gallery:
   • In the search bar at the top, search for "Azure Compute Gallery" and click on "Create."

• Fill out the required details:
  • Subscription: Choose the subscription in which the gallery will be created.
  • Resource Group: Select an existing resource group or create a new one.
  • Gallery Name: Provide a unique name for the gallery.
  • Region: Select the region where the gallery will be created (it’s recommended to place the gallery in a region close to where most of your VMs reside).
• Click on Review + Create and then Create to deploy the gallery.

---

Step 2: Prepare the Virtual Machine (VM)

Before capturing an image, ensure the VM is configured and ready to be captured. It’s essential that the VM is in the correct state.

1. Access the VM:
   • In the Azure Portal, navigate to your VM, ensuring it is in the desired state, with all necessary applications, configurations, and updates installed.

1. Configure the VM for Image Capture:
   • If you want the VM to be specialized, it means the machine has unique configurations like specific user settings, applications, and data. The specialized state is typically required if you intend to use the image for subsequent VM deployments where those customizations should persist.

• Specialized vs. Generalized VM:
  • Specialized VM: This is a VM that has been fully configured with custom settings. It is generally used when the machine will act as a template for an environment that needs to maintain specific configuration or software settings. It is not suitable for creating images that will be used across multiple tenants or environments.
  • Generalized VM: This is a VM that has been generalized to remove any unique configurations like user accounts or specific machine-specific data, making it ideal for creating reusable, uniform VM images that can be used to create new VMs. A generalized VM is prepared by running the command sysprep in Windows or deprovision in Linux.

---

Step 3: Capture the Image of the Specialized VM

1. Ensure the VM is in Specialized State:

   • If the VM is configured correctly, and you want to capture it as a specialized image, leave the VM in its current state. If it is generalized, refer to the next section for instructions on how to generalize it.

2. Capture the Image:

   • To capture the image of the VM, follow these steps:
     • In the Azure Portal, go to the VM and select Capture from the top menu.

 - Name your image, and select the **Compute Gallery** you just created.
 - Choose the **Image version** (this is how you will track different versions of your image).

 - Make sure to check the box that indicates the VM is **specialized** and **not generalized**.

 - Confirm that you understand the implications of capturing a specialized image.
 - Click **Create** to capture the image.

The VM image is now stored in your Azure Compute Gallery and can be used to deploy new VMs with the same configuration.

---

When to Generalize a VM

While we focused on capturing a specialized image, there are situations where you may want to generalize the VM instead. The primary use case for generalizing a VM is when you need to create a reusable, standardized image for scaling purposes.

1. Generalizing the VM: To generalize the VM, use the following process (for Windows VMs, the sysprep tool is used, while Linux VMs use the deprovision command):

• For Windows VMs:

  • Open the Command Prompt inside the VM.
  • Run the following command:

     sysprep /oobe /shutdown /generalize
  

  • This command prepares the VM by removing unique machine-specific data and settings. The VM will shut down once the process is completed.

• For Linux VMs:

  • Open the terminal inside the VM.
  • Run the command to deprovision the VM:

     sudo waagent -deprovision+user -force
  

  • This will remove any user-specific data and settings, preparing the VM for image capture.

1. Capture the Image After Generalization:
   • Once the VM is generalized, you can proceed with capturing the image by following the same steps we used to capture the specialized image, but this time, the VM image will be available for widespread deployment and will not contain unique data.

---

Step 4: Use the Image from the Azure Compute Gallery

Once you have captured the image, whether specialized or generalized, it’s available in your Azure Compute Gallery. You can use it to deploy new virtual machines within the same region or across different regions.

1. Deploying VMs from the Compute Gallery:
   • In the Azure Portal, go to the Compute Gallery and select the image you captured.

• Choose Create VM to initiate the deployment process.

• Follow the prompts to configure the VM settings, such as size, network, and storage.

---

Conclusion

Creating an Azure Compute Gallery and capturing a VM image is a vital part of managing and scaling virtual machines in Azure. Deciding whether to generalize or specialize the VM depends on your use case. Specialized images are ideal when you need to maintain unique machine configurations, while generalized images are more suitable for creating multiple identical VM instances. Azure Compute Gallery provides an efficient, organized way to manage these images, improving the overall efficiency of your cloud infrastructure.