Disaster Recovery and the Darker Side of the ‘Undo’ Function

There are times when you wish you could undo what you just did. Sometimes, you can’t. Financial investments, office reorganisations and even that too-hasty email you sent often cannot simply be reversed. With IT on the other hand, it’s a different story. From individual PCs to corporate data centres, the ‘Undo’ function has become a standard feature of many computing systems for making errors and problems disappear. As little as one mouse click may be enough to turn back the hands of time and begin again as though a mistake had never been made. But is this disaster recovery capability the magical solution it is often made out to be?

In many instances, the undo function, where provided, works well. The individual catastrophes that have been averted in PC applications such as spreadsheet and text-processing thanks to a quick ‘control Z’ or its equivalent are countless. Likewise, the restore point functionality in operating systems like MS Windows have helped many a user get back to normality after an ill-considered reconfiguration or software installation. At a higher level, the same snapshot restore and versioning principles also apply to corporate IT servers and databases.

However, even within the rational universe of information technology, ‘Undo’ does not always undo. Possible causes of undo failure include shortcomings in the software itself and malware that is smart enough to hide from the undo function. You could always reformat your entire hard disk to eliminate infection, but that’s no longer ‘undo’ – it’s ‘obliterate’. Unconfirmed assumptions and untested disaster recovery procedures further compound the ‘Undo’ problem. Users must be told the truth. The undo function is not infallible. Smart disaster recovery planning will however recognise the risk however small of failure and take steps to provide alternate solutions; or at least reduce the impact of any serious ‘Undo’ failure.