Although administrators have more VM backup and restoration options than in traditional environments, there are...
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
First, administrators need to be aware of the status of any buffers or cache related to their applications. It may be necessary to account for this temporary data before taking a VM backup snapshot.
"Your SQL database is running in a VM, and you've backed up the VM," said Greg Schulz, founder and senior analyst at the StorageIO Group, an IT infrastructure technology analyst and consulting firm in Stillwater, Minn.
If something happens and that SQL server hadn't committed all its transactions, then you've got an inconsistent database, Schulz said.
"The virtual machine might come back up, SQL might come back up, and you find that table corrupted," he said.
Schulz said that the only option then is to recover from the last known good backup and restore the damaged database from that point. This kind of synchronization may be performed automatically when quiescing some applications like SQL. Also, tools like VMware Consolidated Backup may capture such buffered data as a normal part of the tool's operation.
In either case, the goal is to ensure that a VM and its application are in a fully stable state when a VM backup snapshot takes place. Proof-of-principle testing can identify possible synchronization problems before rolling out critical VM backup snapshots.
Another issue is restoration. Snapshots are highly effective, but an entire VM is captured as a single contiguous file. Consequently, VMs have to be restored as a single complete entity, so snapshots don't lend themselves to single-file or folder restoration. Organizations often rely on operational backups for granular or noncritical data recovery.
"Operational backups are the kind of backups that you use to restore the CEO's one email that he deleted or that Word document that got overwritten," said Dave Sobel, the CEO of Evolve Technologies, a tool provider in Fairfax, Va. "They're the kind of backups that you need for rapid recoveries of small amounts of data."
Some organizations achieve file-level granularity by performing VM restores to an auxiliary server and then simply extracting the files that need to be restored for a specific user. Another alternative is to select VM backup and restoration tools that can "see inside" a VM snapshot and restore specific files or folders as needed. Symantec Backup Exec tools tout exactly this type of granular access in a virtual environment.
A conventional bare-metal backup protects the OS, drivers and every nuance of the physical server's configuration, along with the application and its data, but a VM backup does not include the underlying hypervisor and OS.
Although a VM is a complete and mountable machine file, it is not "bare metal" in the traditional sense. The destination server must already have an operating system and hypervisor in place. The OS and hypervisor must be backed up as a separate step.
"What's your backup strategy for the physical host?" asked Sobel, adding that some organizations might opt to forgo the OS/hypervisor portion of a VM backup, electing instead to restore the bare-metal portion of a virtual server from original media.
About the author
Stephen J. Bigelow, a senior technology writer in the Data Center and Virtualization Group at TechTarget Inc., has more than 15 years of technical writing experience in the PC/ technology industry. He holds a bachelor of science in electrical engineering, along with CompTIA A+ and Network+ certifications, and has written hundreds of articles and more than 15 feature books on computer troubleshooting, including Bigelow's PC Hardware Desk Reference and Bigelow's PC Hardware Annoyances. Contact him at firstname.lastname@example.org.