Making the Grade: Guidelines to Ensure Solid Disaster Recovery with Testing

By Dan Timko

Dan Timko

While managed service providers and enterprise IT pros backup and restore data all the time, few have ever had to recover from a true disaster. Disaster recovery (DR) is a different, more complex beast than restoring an application or discrete data from backups. Especially given the limited experience most have in these situations, taming DR requires regular and comprehensive testing. Unfortunately, testing, if it happens at all, is often partial and slipshod. A study from Spiceworks showed that nearly a quarter of organizations do no DR testing at all.

Without thorough testing, a service provider or enterprise can suffer serious damage to their business and brand should it fail to recover promptly from a massive cyberattack, fire or other disaster. Essentially, all DR efforts are wasted without regular testing, because without a restoration plan that’s been tested to show it works, applications may not restore to a running state, data could be infected with malware and IT may not restore applications in the proper order so that dependencies are met.

These aren’t problems you want to learn about as employees sit idly by and the C-suite anxiously paces.

Why It’s Best to Test

Assume recovery can fail. Why? There are many scenarios where they fail, but, for example, if backups are corrupted due to a storage issue, recovery is doomed — and that’s an issue that would be exposed with testing. But that’s just the start of what testing can unearth, and there are related shortfalls that create vulnerabilities, including the following:

Guidelines for Solid DR Testing

Define what needs to be tested. IT should first evaluate what must be recovered, assess each application’s criticality to establish priorities, and identify interdependencies. There are more complex operational-centric aspects that may also need testing, including:

Testing Models

There are three primary purposes when testing DR: Test the plan, test the people, test the technology. Each of the following models …

… covers at least one or more of these elements:

Making the Grade

When finances get tight, decision makers, and sometimes even IT leaders, will cut DR and count on the unlikelihood of a catastrophe working in their favor. It’s a classic case of being penny wise and pound foolish when you consider how even minor downtime can result in missed sales, unhappy customers and blemished reputations, all to save a few bucks in the short term.

MSPs and IT need to convince decision makers that disasters can and do happen. If they’re not swayed, remind them that at the start of 2020 no one was predicting a global workforce would soon go remote and with only a few days’ notice. And really, with this new paradigm, IT must now re-evaluate and adjust strategy so that business keeps going under any circumstance.

Testing is more important than ever, and by doing so, you can ensure that come what may, your DR efforts will make the grade.

Dan Timko serves as chief strategy officer for cloud backup at J2 Global, which includes J2’s KeepItSafe, OffsiteDataSync, Livedrive and LiveVault businesses. Prior to joining J2, Dan co-founded Cirrity, a BaaS and DRaaS provider that was sold to Green Cloud Technologies in 2017. Follow him on LinkedIn or @dan_timko or @j2global on Twitter.