What is likely to cause downtime when changing the size of a running Azure virtual machine?

Changing the size of a running Azure virtual machine is likely to cause downtime, particularly when scaling the VM to a different size, such as from D4s v3 to D8s v3. This process typically requires the virtual machine to be stopped and deallocated before resizing can occur. During this period, the VM is not operational, leading to an unavoidable downtime as the VM needs to be completely shut down to make the necessary adjustments to its resources.

In contrast, activities such as adding a managed disk, enabling Desired State Configuration (DSC) Management, or adding an extension can often be carried out while the VM is running, thereby minimizing or completely avoiding any downtime. These actions involve changes to the configuration or features of the VM, but they do not necessitate stopping the VM itself, which is why they are less likely to cause an interruption in service.