There are strong cases to be made for virtualization:
- Reduction in power costs
- Better utilization of hardware resources
- Easy transition to new hardware in the event of a host failure
The list goes on. But one application of virtualized service that is overlooked in many cases is a virtual system backup strategy.
It is very easy to take snapshots of virtual systems, or stop the host services and take actual copies of the virtual machine files that can then be easily ported to alternate hardware. What happens, though, when the system itself becomes unstable through system file corruption, or infected with a virus?
Perhaps the virus itself infected the machine a month ago, corrupting data along the way, but the symptoms were not noticed until well into your snapshot cycle. With the virtualized operating system compromised, and no way to know how far back the damage began, you cannot simply restore a copy of the system from last night.
While you may have snapshots going back months, if this virtual machine is acting as a database server, a file server, a web server, or anything else that end users may be updating, restoring a month old machine could lose hundreds of man hours of work.
For all virtual environments that contain dynamically updated data, a standard backup strategy should be considered that will protect the business from data loss. The machine snapshots can and should be used to quickly recover from a host hardware failure, but to properly protect your data, ensure your data can be recovered even if the machine itself cannot.