Test Restore

Every good backup and recovery plan involves testing. It is extremely important to perform test restores from your backup media occasionally. This is especially important with tape media because tapes wear out and tape drives get dirty over time. CDs can get scratched or mishandled which could make them impossible to read. Another problem with CD and DVD media are compatibility issues. I have burned CDs and DVDs on one drive that cannot be read on another. The compatibility problem is not as common as it was a few years ago, but you don't want to find out you have a problem when the only copy of some valuable data is on a CD that you can't read. Media problems aren't the only reason to occasionally test your restore capability. You could very well have issues with the restore functionality or security issues with the operating system.

Online backup systems are no exception and should be tested as well. Most online backup service providers protect you from being directly exposed to media problems by the way they use fault tolerant storage systems. While online backup systems are extremely reliable, you would be well advised to test the restore side of the system to make sure there are no problems with the retrieval functionality, security or encryption.

I have been involved in some very comprehensive disaster recovery tests where a team of IT professionals would bring up an entire data center at a disaster recovery center, using only backup media. This is the ultimate test. Most large corporations are required by auditors to perform such exercises periodically. The expense of maintaining and testing a comprehensive disaster recover plan can be substantial. Smaller businesses can't afford the full blown disaster recovery plan so they rely on their data backups and off site storage plans to save them in case of a catastrophe.

If you are the typical small business operator or home user, you can perform a few simple tests to ensure that your backup files will be available when needed. First of all I am assuming that backups are being done on a regular basis. Probably not a good assumption based on conversations I have had in recent months, but that is a topic for another article. It suffices to say that you must be making backups if you expect to recover anything.

To perform a simple restore test of your backup system. Follow these simple steps:

1) Rename one of your document folders. Don't delete it because the restore could fail.

2) Now see if you can find the correct tape or media that contains your most current files. This is an important step in the test because it can be difficult to locate the correct media if you don't have a good system of keep track of what files are on which media and the dates they were backed up. Online backup systems have a huge advantage here because they usually keep a nice need catalog for you of what was backed up and when, and you don't have to locate any media.

3) Use your backup system to restore the folder that you renamed in step one. Restore it back to it's original location.

4) Count the files in the folder that you renamed and in the folder that you restored. In Windows you can right click on each folder and select properties. Make sure the counts match, or make sure you can reconcile any differences.

5) Open a few of the files and make sure they are accessible. Be sure to open them with the application that normally uses them. And make sure you can browse the data, update the data or whatever you normally do. It is best to open several files of different types, pictures, mp3s, documents and some large and some small.

Most operating systems have utilities that you can use to compare files and directories. I will not go into those here because there are many options and many ways to get false alarms with those utilities. If you are computer literate, then by all means, go ahead and use the utilities to do a comprehensive comparison. I promise to write a future article that goes into the use of some of these tools.

I caution that the above is not a comprehensive test, but is it more than most people are doing now and can expose problems that could cause some severe headaches if not discovered.

As for the frequency of testing your restores. That depends on several factors; I would recommend twice a year at a minimum. If you make changes to your hardware, software, or backup program; then you should test after each round of changes.

You are backing up your files so that you can recover in the event of data loss. By the way, data loss never occurs at a convenient time. It will almost always occur when you don't have time to deal with it, when you are in a hurry to meet a deadline, or when you have something more important to do. You will want to be able restore your files quickly and get back to doing what you had planned to do with your time. Losing your data is bad enough and will put a serious cramp in your schedule, the last thing you want to find out is that your backup is no good. Do yourself a big service and find out if your backup is no good at a time when you have not lost your data.

No comments: