In the world of software development most testers know the phrase “Happy Path.”  Happy Path is when you have a very well defined test case that has no unusual events. If you test your software on the Happy Path, it works, and everyone is happy.  Unfortunately, just because software works on the Happy Path, doesn’t mean it will work in production environments, under heavy loads, and under a variety of fault conditions.

Too often, today, disaster recovery testing is done on the Happy Path.  If you tightly control the disaster recovery test plan, you can prove to your executive team and your auditors, that you can recover your data, your applications and your business processes.  But get off the Happy Path, and the probability of recovery quickly approaches zero.

The real world is filled with randomness and unpredictable events. Often, multiple bad events occur in close succession or even simultaneously.   But because organizations test their recoverability on the Happy Path, they delude themselves into thinking that they can actually recover in the event of a real disaster.  I think it’s time that we all get off the Happy Path and start building disaster recovery plans that enable organizations to survive the unpredictable and the unplanned.