How to Best Protect your Virtual Environments from Data Loss

Monday, November 14, 2016 by Michael Nuncic

After more than a decade since the release of virtual environments, a new study by Gartner states that the virtualization market is now mature. What the experts found is that even though the worldwide virtualization market has yet again increased by almost 6 percent to reach $5.6 billion USD, they do not expect large gains in the future. What they found is majority of large companies have already virtualized their servers to the highest degree – with many organizations having server virtualization rates that exceed 75%. Virtualization is now far from being a niche technology and has moved into the mainstream.

With so many companies using virtualization in their IT environment, it’s no wonder that data recovery experts have experienced a significant rise in projects for virtual environments and machines over the last few years. Virtualization, in combination with other advanced server and storage technologies like deduplication, hyper-converged storage or RAID often results in layers data recovery experts have to dig through to finally reach and retrieve needed files or data.

To avoid unnecessary data loss in virtual environments – here are six important tips on how to protect your files:

1.  Use the right backup software for your virtual environment

There are various backup software solutions on the market which can be used with virtualized files. Some can be used both with VMware or HyperV solutions, but this is not the most important factor you should consider. The most essential factor is how much time is needed to recover virtual machines (VMs) out of the backup and the VM templates. Additionally, good backup solutions for VMs are able to mount the backup while the files are transferred back to the main host system.

2.  A snapshot is not a backup

Always create backups if you want to make sure that everything can be fully recovered in the event of a failure. Additionally, you should create snapshots if your changes are very important and you don’t want to lose any data, so you can cover the time frame between the last and the next backup. If you use snapshots, don’t try to build snapshots that build on each other. Some “experts” manage to create 6-7 layered snapshots, which not only reduces performance, but makes them more susceptible to error. And the damage of a VMFS or the failure of a physical server cannot be handled by using a snapshot.

3.  Don’t save backups and running VMs on the same storage

If you save your backups on the same hard disk or storage space where your active VMs are located, you risk ending up with a total data loss. If a backup fails and a VM is active, it’s likely that it will overwrite the backup. In order to prevent this from occurring, always store your backup VMs inside and your active VMs in a different place. It also makes sense to make multiple backups and store them on a different local server/hard disks, whether in the cloud or on tape. Having two additional backups can help you in the event of a data loss.

4.  Don’t try to mix several virtualization solutions

If you use VMware and HyperV virtualization solutions in the same environment, it’s not unlikely that you’ll experience weird outcomes which can lead to data loss. If, for example, a VMware system on one SAN is expanded onto a second SAN which is used by HyperV, the outcome can be a severe data loss on the second SAN. The different layers that the virtualization tools create behave totally different when they are linked and/or expanded. So better keep your virtualization simple and stick to one virtualization solution. Too many layers of complexity will definitely make your system crash someday and will make data recovery an even time consuming project – if possible at all.

5.  Only use technologies available with your virtualization solution wisely

There are several technologies available that have an impact when used with virtualization. One example is thin provisioning. Thin provisioning in its simplest form means that the only storage space being used is really needed in the current moment. When additional space is needed, free space will be allocated freely. But this comes with a side effect: In case of a virtual data loss, the system must be stopped immediately, otherwise it’s possible that when other virtual hard disks are running on the same disk, this space will be used to store new data on the available free space. Subsequently, data recovery can then become more difficult. With this in mind, it’s always a good idea to first - think before you use a complex technology and second – better use a more simplified approach. This can help when it comes to data recovery and the time and budget needed later on.

6.  Think and plan before you use VMs

Always keep in mind that virtualization and virtual machines are not error-free and are as likely to failure as every other technology. So before you create a virtual environment for sensitive applications, think first. Some applications have a high input/output (I/O) rate and are therefore better fitted for physical server environments. Planning ahead regarding virtualization is the key, when it comes to preventing data loss. One of the main data loss problems with virtualization is that the setup of the virtualized server and storages is insufficient. Missing documentation of VMs and virtual servers being built and connected with applications and sensitive data included can make data recovery a costly and time-consuming task.

But don’t fool yourself! Even if you implement those six basic tips, you can still experience a severe data loss involving virtual environments and machines. Since every virtualized IT environment has its own advantages and disadvantages of how it works best, the administrator in charge should be fully aware of every detail on how the system works. Even better, they should have learned the specific manual on how to react in case of failure, data loss and data recovery by heart. If you are not fully aware of what to do and why, you should not try to recover virtual machines yourself. Better consult a specialist like Ontrack – it pays off in the end!