Please note that when you scale up/down a VM, it is restarted prior to allocating new resources to it. As it is difficult to alter provisioned CPU and memory for a working machine, it is highly possible that such an operation will restart your workloads or pause them. This is why you should do that with caution – wait for a window of opportunity or low-peak hours to limit users affected by the scaling operation.
Remember that a scaling operation may fail due to various reasons:
- Temporary data center capacity quota
- Service outage
- Reaching limits on your subscription
Take these into account when you plan to make any changes to the production subscription to avoid disruptions to your services.
Soon, you will see that scaling is one of the most common operations you can perform when working with cloud services. In the next section, I will show you how to configure another important feature of cloud services, which is monitoring for VMs.
