“Safe and Reliable” are two words frequently used when articulating the features and benefits of computer back-up systems. But just how safe is “Safe” and just how reliable is “Reliable”?

After installing a reasonably expensive automated back-up and recovery system on our network, we followed the installation instruction guide and did a complete back-up followed by a complete system restore.

The successful completion provided a sense of confidence that, should disaster strike, we were only minutes away from initiating a restore of the data and being back in business.

For years we were under the delusion that our automated back-up and recovery system was functioning correctly. After all, there were no device errors reported through the sys-admin tools. That, coupled with our continuous cycling and replacing of the backup tapes, strengthened our belief that our “Safe and Reliable” system would be there at a moment’s notice. NOT SO!

Recovering from a disaster is not the time to discover that the “recovery side” of the automated back-up and recovery system is broken, and may have been broken for months. Since we had not needed to recover data–nor practiced a full disaster recovery program–the recovery function of the system had remained dormant and dysfunctional for months. The “glitch” as we discovered later was due to one of those many automatically applied operating system updates that knocked the “stars out of alignment”. Something that would not be reported as a device failure but quickly sidelined our efforts to be back in business quickly.

So lesson learned: test both the back-up and restore functions – regularly. One efficient method to do so is practiced by many Spitfire Project Management System-installed clients. By deploying two sites, a live site and a training/test site (recommended by Spitfire at no charge), the live site data is backed up and then restored to the training/test site on a regular basis. This has several benefits: a) it validates that the hardware needed to back-up and restore are working correctly, b) it verifies that the back-up media is handling the data properly, and c) it provides a non-destructive training site, emulating the live site with current project data, so that user training does not hamper or impact the live site.

A simple and effective solution for hungry dog-gone-computers.