When was the last time you tested your Disaster Recovery plan?

by | Sep 15, 2017 | Infrastructure | 0 comments

Creating a disaster recovery (DR) policy takes a lot of effort.  You need to consider all of the situations that could occur, identify your most critical systems and processes, define requirements, and document the plan.

However – have you reviewed all of the above recently?  Disaster recovery is a process that needs to be refreshed and challenged to meet up with current business requirements and changes to the technology landscape.  In this blog, I share an approach for testing and items you should consider.

Creating the test plan

Disaster recovery testing can vary widely depending on the size and scope you want to review.  All of the different disaster recovery scenarios for which you have planned can dictate the complexity of the testing.

  • What constitutes a disaster that requires a fail over?
    • Do we fail over if just a specific system or resource fails, and if so do we fail everything over or just that resource?
    • Do we fail over if we are not able to access our building, but the resources are still in there and running and we have access to them remotely?
    • Do we only fail over if the entire building is somehow unavailable for an indeterminate amount of time and all of our resources are down?
  • What is the timing of when we fail over?
    • Do we fail over as soon as there is a failure?
    • Do we try to recover our primary site for a set amount of time before declaring a disaster?
  • Who will be available to implement a failover?
    • Do we have resources available at our failover location to initiate the failover procedure?
    • Do they have the right access to perform the fail over?
  • How do we fail back?
    • Do we have to rebuild infrastructure entirely?
    • Do we have the ability to just initiate our replication technologies in the reverse direction in order to bring our primary site back online?

These are just some of the questions that can help you create the different tests for testing your disaster recovery plan as you consider various scenarios.

You can make your disaster recovery plans as intricate and complicated as you would like, but the testing of these plans will usually be just as intricate and complicated.  That will also have a direct correlation to the amount of time and cost associated with testing the plan.

Getting outside expertise and assistance

Not all testing plans are created equal and there are so many options when it comes to testing.  Some organizations are not comfortable testing with production data.  On the flipside, some organizations may feel like it is not a valid test unless they actually fail over production data to their disaster recovery site and are able to fail it back to their production site.  Some tests may involve failing over in the way that enables testing of all functionality from end to end to validate that everything is fully available just as it is in production.  Some just replicate everything into a virtual environment, proving everything is able to be brought back online and that the data is accessible.

Because of this complexity, it can be helpful to bring in external help to find the right strategy for your organization. At Peters & Associates, we help create the test plan to fit your needs by guiding you in exploring all of the questions that will lead to the plan that is most appropriate for your organization.  We can also help develop and implement a testing methodology to test the plan for the different scenarios you have outlined.

Leveraging technology for DR planning and testing

New offerings from companies like Microsoft and their Azure Site Recovery feature change the game when it comes to disaster recovery planning and even testing.  With Azure Site Recovery, you have the ability to not only fail over physical machines, but also virtual machines running Hyper-V or even VMware.  In addition, there are capabilities of even failing back to your original infrastructure, even if that original infrastructure is running on bare metal or non-Microsoft hypervisors such as VMware.

We have helped many organizations figure out how to leverage these new options to update their DR and testing plans to be more effective and efficient.  These technologies can provide peace of mind at a reasonable price point.

I admit – I don’t even have a disaster recovery plan!

With all of the other priorities in your organization, you may not have formally created a disaster recovery/business continuity plan yet.  Don’t worry – we can help you create, maintain and test that plan.  We can also help create or locate the place to implement your plan.  That could simply mean coming up with a list of parts to order in the event of a disaster so that you could restore all of your tapes to over a week.  It could also involve finding a colocation site and implementing replication technologies that would allow you to failover in a matter of minute.  We can help with those scenarios and everything in between.

To learn more, contact us at info@peters.com.  We are happy to help!