The world runs on data. Your bank account isn’t money, it’s data, sitting in a (hopefully) secure server that determines how much you can spend. Your customer’s orders are just data, sitting in your system, keeping a faithful accounting of inventory stock and revenue. Your business, your life, are all just bits of ones and zeros floating around in cyberspace. And it all runs along smoothly. Until it doesn’t.
With my background in eCommerce, two things have always had a special importance in my world. Downtime and data loss. Downtime can mean a drastic loss of revenue and productivity. Data loss can be a business-ending catastrophe. And the best way to avoid this apocalyptic event is with a strong data backup and recovery strategy.
Your Backup and Recovery Plan
While all businesses and networks are different, the road to a great backup and recovery plan is the same:
- Identify key data points
- Define crucial system environments
- Create a backup and recovery plan
- Verify the plan works
- Practice the recovery plan
Identify Key Data Points
If you are gonna backup your data, make sure you get it all, or at least everything you can’t live without. And if you’re in business, that pretty much means everything. If you are like most businesses, you have a file server full of spreadsheets and important documents, several application servers with your financial, inventory, and customer management software, and a host of other important files scattered around your office computers. It’s important to take a solid inventory of all your data points and consolidate them. Enforce office policies mandating the use of shared drives, dropbox, Sharepoint, and other central file locations that are more easily backed up. Make sure you have a handle on all the files and data your business needs to run.
Define Crucial System Environments
Having all your files backed up and secure is great, but often your most important data relies on applications and software environment to run. These are things like your databases, accounting software data, and inventory and order system databases. Being handed a backup of this is about as helpful as me being handed the ingredients to an amazing Emeril Lagassee dinner recipe – I don’t have the know-how and tools to do anything with the ingredients. So just as important as the data is identifying the server environments, applications, and configurations needed to properly handle your data and get you up and running. Remember, the end goal of the recovery is to be able to get up and running as quickly as possible, and that won’t happen if you can’t recreate your server environments.
Create a Backup and Recovery Plan
Sounds kind of obvious, right? But you’d be amazed at how often companies have steps 1 and 2 down, but have no plan in place to do anything with their data. Your plan needs to define exactly how you are going to get from Point A (oh no, we’ve lost everything) to Point B (whew, we’re back up now). This includes identifying where you are restoring to (Virtual Machine? Server? Located at a data center? Cloud Provider?), how you’ll get your data there (uploading 1 TB of data over a T1 will take quite a while), who is the key person responsible for the recovery process, what support they’ll need (VPN connections? Server OS guru? PO from finance?), and the exact steps involved in recreating the environments and getting the data back online. This is no small matter. Don’t assume anything, including hardware. Imagine you got a call at 3am that your office burned to the ground. This is the type of disaster you need to be prepared to recover from, as well as a simple server failure in your office.
This is where the rubber meets the road. Put your plan in place. As mentioned in our page on data backup and recovery services, your plan should be as automated as possible. Not in a “I have an Outlook reminder set for 3pm on Fridays” type of way, but in a true, run behind the scenes, no one needs to touch it way. There is software out there that does a lot of this, but often customized scripting is needed when dealing with cloud backups, VPNs, multiple office locations and other circumstances. Whatever technology you choose, make sure it is backing up multiple times a day, ideally to multiple locations. The worst type of backup plan backs up to a hard drive located in the same building (or worse, the same server). While this is often a great safeguard against server and hard drive failures, this won’t save you from flood, fire, or theft.
Verify the Plan Works
Of course you’ll do this as soon as the plan is implemented. “Sweet! The backups are there!” Then six months later, no one has checked the backups and realized that when a server was fixed, the drive letter changed and the backup plan stopped executing.
Verification means regularly verifying the data is being backed up as scheduled. Additionally, have the backup processes send failure notification emails that notify of successful and failed backups.
Practice the Recovery Plan
The best laid plans of mice and men, often go awry (Robert Burns). Don’t wait until disaster strikes before you attempt a recovery. Everything goes better the second (or third) time around. Practice the recovery. Schedule it in advance, get the key stakeholders ready, and start from scratch with a blank environment. You’ll run into snags, no doubt, and that is the point. Find the bottlenecks now. This is the fire drill, not the fire. Fine tune your action plan, necessary resources, and realize the things you’ve forgotten. Then, a month later, do it again. After a couple or three go ’rounds, you’ll be confident in your ability to recover from any disaster.
At SystemGo IT, we hope this helps you to weather the eventual storm. We welcome your feedback and suggestions in the comment section below, and if you have any questions or need some help, please contact us.