A few weeks ago Microsoft announced the release of the VM Depot, a community-driven catalog of VM Images. Deploying a Virtual Machine with an image from the VM Depot was pretty easy:
- Download/Upload the VM image yourself.
- Use the azure-cli to do all the work
But it looks like the VM Depot now seamlessly integrates with the Windows Azure Portal. Let’s take a look at how easy it is to create Virtual Machines based on VM Depot images from within the portal.
Create the image
If you browse to Virtual Machines in the portal you’ll see the new Browse VM Depot option under the Images tab:
Next you’ll be able to browse the images available in the VM Depot:
After selecting an image you will need to choose in which storage account it will be dropped:
Wait a few seconds for the copy to complete. Once it’s done you’ll see that the image appears in the Images tab with a Pending registration status:
Click the Register button to register the image which will change the status to Available.
Creating the Virtual Machine
That was the “hard” part. Now in order to create the Virtual Machine you’ll need to go to the Virtual Machine Instances tab and choose Create a Virtual Machine. Choose to create it from the Gallery. In the Gallery you’ll see the My Images tab and this is where you’ll find the image:
Select this image and go through the wizard (enter the name of the machine, username, password, …). The provisioning process will start and after a few minutes the VM will be ready. This means you’ll be able to connect to the VM through SSH (myvmname.cloudapp.net port 22):
If you want to connect to the Solr dashboard you can also add an endpoint for port 80:
And here is the dashboard:
This is really great. In just a few minutes I’ve set up a fully functional Solr deployment, which I can use in my applications to offer an improved search experience. If you want to use this in production I suggest you also look at Virtual Networks. This feature allows you to connect the VM to your applications (running in Cloud Services/Virtual Machines or even on-premises) without having to expose it to the internet.