When did you last test your disaster recovery strategy?

This post is a departure from our normal Q&A style because I think this is a very important question you have to ask yourselves.

Related: The Smart DBA's Guide to SQL Server Disaster Recovery, Part I

During our first class of the year a couple of weeks ago, the most recent large earthquake struck New Zealand. One of the students (Martin) was from Christchurch and almost flew home that day, but after finding out that family and work were safe, decided to stay.

Just this morning an incredibly powerful earthquake hit Japan, with an even more destructive tsunami causing devastation.

Apart from the appalling loss of life in both cases, there’s also the economic impact to consider from companies being unable to conduct business.

Which brings me to my question for you: When did you last test your disaster recovery strategy? It’s all very well having a strategy, but how do you know it works?

Conduct Regular Testing

You have to conduct regular testing so you can tell whether the assumptions you made when the strategy was formulated still hold - such as the time it takes to perform a bare-metal restore fits within the Recovery Time Objective.

Later in the week in our class, Martin gave us a short lecture about how his business had reformulated their disaster recovery plan after the previous large earthquake hit New Zealand. He raised some interesting points like:

  1. What happens if you can’t physically get to work? In one case a road had buckled and so couldn’t be driven on.
  2. What happens if the phone service isn’t working? What about if the cell phone service isn’t working?
  3. Which is more important to people in a disaster – making sure the company they work for is up and running or helping out friends and family? Most people will choose the latter – it’s human nature. How does that affect the disaster recovery strategy?

I wrote a longer blog post about disaster recovery testing a few years ago – see Importance of testing your disaster recovery plan.

Edit: And I’ve just kicked off a survey on my personal blog about testing DR plans and whether they anticipate human nature. I encourage you to respond (it’s anonymous) – we’ll be surprised by the results I think. See Surveys: DR plan testing and considering human nature.

Allow for Human Nature

Events like these should get us all thinking about whether our disaster recovery strategies are adequate, and whether they allow for human nature during a disaster.

Stay safe.

Please or Register to post comments.

What's SQL Server Questions Answered?

Practical tips and answers to many of your questions about SQL Server including database management and performance issues.

Contributors

Paul S. Randal

Paul Randal worked on Microsoft's SQL Server team for nine years in development and management roles, writing many of the DBCC commands. Randal was ultimately responsible for SQL Server 2008'...

Kimberly L. Tripp

Kimberly L. Tripp has been working with SQL Server since 1990, and she’s worked as a consultant, trainer, speaker, and writer specializing in core SQL Server performance tuning and availability...
Blog Archive

Sponsored Introduction Continue on to (or wait seconds) ×