Disaster Recovery Is More Than Just Backup
Are you prepared for an office disaster? If a fire, flood, or robbery irrecoverably damages your computer equipment, will you be able to get up and running again? The first requirement, of course, is that you have all essential data backed up, but that’s not necessarily enough by itself.
One business in four that closes because of a disaster never opens again; some estimates are even more pessimistic. Those that do may suffer significant loss of business, with some customers never returning if restoration of service takes too long. A good recovery plan needs to include concrete steps for recovering all data and resuming operation, with a realistic estimate of how long it will take.
To be useful in case of a disaster, at least one backup copy has to be offsite. A backup drive in your office is useful for recovering corrupted files, but if your computers burn or are stolen, it’s very likely the backup will be too. Use a reliable backup service and make sure you know how to get files back from it. Choose a service that has a good reputation and encrypts all data in transit and in storage.
You should also keep a disk image backup of your primary drive. Simply restoring files to a new drive won’t necessarily replicate your original operating environment. Keep the image backup stored offsite in a secure place. Encrypt this backup too, as with anything important that you store offsite.
Make sure that you have the passwords and keys that you’ll need to perform the recovery in a safe place. If you keep them on the premises, put them in a locked, fireproof box that can’t easily be stolen. If they’re off the premises, make sure they’re stored securely.
Plan for how you’ll inform your employees, customers, and business partners of the situation. If their contact information was all on the lost computers, this could be difficult. Keep a copy of it in another place.
Make a realistic estimate of how long it will take to restore your data and resume operation. First you have to set up a new computer and copy your saved disk image to it. After that, you need to restore all your data from the backup; this might take days, depending on how much you have and how fast your connection is. If you have to restore multiple computers through the same Internet connection, it could take a long time. Finally, you’ll need to verify that everything is working properly before going live.
You’ll need to consider what it takes to catch up with lost time. If you send out reports on the first of the month and a disaster brings your systems down on that day, you need to get those reports out as soon as you recover, not on the first of the next month. Make sure it’s possible to do this.
Estimate what the process will cost. You’ll have to buy new computers, replace furniture and supplies, and weather an interruption in revenue. Hopefully insurance will cover most of the loss, but you need enough money in the meantime to get through.
However good your plan sounds, you need to test it to make sure it works. A full test, including failover and restoration of data, may not be feasible, but you should test as much of the plan as reasonably possible. To avoid disruption, you should break the tests down into units that can be tried separately. For example, shut off the main circuit breaker to the office late at night, then get everything back online. Record how long the recovery procedure takes and what problems you encounter. Update the procedure if necessary.
Keep your recovery plan up to date. Even the smallest offices regularly add or replace equipment. Make sure that the plan covers all of it.
To learn how we can help you with business continuity, disaster recovery, and other IT needs, please contact us.
Planning & Consulting
Professional IT Management
Engineering & Support