Five common VM backup headaches that are easily prevented

By Guest on 19 November 2015
Five common VM backup headaches that are easily prevented

We’ve all been there. It’s 2:00am, you have a server flat on its face, and you’re having issues getting your backups to restore properly. It’s more common of a situation than most people realise, and 2:00am with a down server is not the time you want to come to that realisation. Talk about a headache.

A backup routine for IT departments is so common and unquestioned these days that many fall into the habit of complacency or ignorance. Either something is missed during the configuration of the backup routine, because backups are ‘as simple as clicking a check box’, or proper testing of that backup routine is not done on a regular basis to insure that said backups are in proper working order.

Whatever the reason for that 2:00am phone call, let’s cover some backup related headaches that can be prevented when working in virtualised environments.

1 - You store backups with production data

In today’s storage ecosystem we have all these fancy storage area networks, network attached storage heads, and Converged storage arrays that serve as a centralised storage repository for production data. The problem that some organisations run into when working with backups in a virtualised environment is ensuring that the backups are NOT stored in the same location as your production data. You can run the backup software from a virtual machine (VM) in your environment, but special care should be given to ensure that the backup data resides somewhere other than the storage that is driving your production systems. If you do store backup data and production data in the same location, just think about that storage device going belly up. How would you recover? 2:00am is not when you want to discover this, so plan accordingly. 

2 - Snapshots are NOT a replacement for backups

Don’t misunderstand. Snapshots are AWESOME, when used properly. It’s a sad fact though that many organisations think they can use snapshots to replace their backup software and then save a penny. Backup software conducts a lot of background operations that snapshots do not. Things like encryption, compression, and deduplication. This is not to mention things like retention policies, replication options and the like, depending on the backup application. Additionally, misuse of snapshots could not only damage the associated virtual machine, they can also cause multiple VMs in your organisation to stop functioning if a snapshot is forgotten, and grows so large it consumes your storage. Choose the right tool for the right job and DON’T rely on snapshots for backups.

3 - Your production site goes dark and your backups were there as well

This builds off the idea that backups should be in a different location to your production data. While this may not be economical for all organisations, it allows businesses to come up with a way to get backups offsite somehow. Whether that is something as simple as bringing a USB drive home every day, it’s still better than nothing. One question to ask yourself when considering how important this is: ‘How would you recover if the location of your production data and backups became a smoking crater in the ground tomorrow?’ Remember, your data = your business, and if the entire site gets blasted by a meteor, you can at least rest well knowing your data also resides somewhere else.

4 - Backups appeared successful, but will not restore

This situation has a VERY easy solution. This can be prevented by regular consistency checks on your backups. If nothing else, just recover a single file from each server once a quarter. Again, 2:00am is not the time to be making this discovery, and if you want a more automated process for this, then there are many ways to automate this process. Sandbox features are built into many of today’s backup applications and issue e-mailed alerts should something fail.

5 - You find out your backup methodology does not conform to your compliancy requirements

It’s the year 2015 and more and more businesses are having to contend with the various requirements put forth by industry regulation legislation like HIPPA and PCI. It’s not a good feeling to find out you’ve been fined because your backups are not stored or conducted in a fashion that conforms to those regulations. Most modern backup applications have mechanisms in place to help with this, such as encryption. Additionally, you’ll want to make sure your configured storage is secure with only the needed permissions to do the job applied and enabled. How you configure both of these items will be determined by the vendor involved in each. Simply something to be aware of.

These are just some of the most common worst-case scenarios that can arise with improper use, planning, and management of your backup solution. As you can see, most of these items could have been prevented with some up front common sense best practices. There is a newly published whitepaper on this very subject if you’re interested in learning more.

Andy Syrewicze is a technical evangelist for Altaro Software, as public speaker, technical author and blogger. His work covers a number of Microsoft solutions.

Number of views (3165)/Comments (-)

Comments are only visible to subscribers.

Theme picker