Critical Data Recovery Problems in the Linux and Unix Environment

by | May 22, 2013 | Business Continuity

Are you protected against Data Recovery Problems in the Linux and Unix Environment?

Linux and Unix are some of the most popular operating systems for enterprise and favored by those who are hardcore about tech. This is because of the speed, flexibility, and customizability that the systems offer for anyone who takes the time to learn their intricacies. Also, they are stripped down and souped-up for machine-to-machine communication; the kind favored within an IT environment. Of course, with great power comes great responsibility, and neither system has a convenient “recycle bin” for you to click on. Data recovery problems are a huge risk and a recovery solution must be in place. Let’s take a look at a hypothetical example.

Say that I am a developer at a financial firm that deals in high volumes of transactions – some of them representing millions of dollars. The software team has dual Redhat Linux development servers as its test environment, on which all QA validation is performed before final deployment to the production servers. Both servers run a third party transaction and auditing platform from a third party – let’s just call them the “XYZ” corporation. The XYZ corporation software keeps all its important data in a directory called /xyz/data. One day I am performing an upgrade on the /xyz software and have to copy two large upgrade files called and into the /xyz/data folder. After the upgrade is done, I decide to delete the files because the partition is low on disk space. I am low on sleep and energy and there are 20 IM windows open on my laptop. With too much haste I aim to fix my data recovery problems by typing the following command:

rm update *
I meant to take care of the two files in one command (rm is the delete command and the update* means to delete everything that begins with update, but the crucial space there means I deleted EVERYTHING in the folder). The content of that folder has all test data and it could take weeks of effort to replace, if it can be replaced at all.

YOU MIGHT ALSO LIKE:  In-house or outsource? Who should manage your business continuity?

I panic and call the X corporation support specialists. They tell me that their system has no built in backup (by the way, thanks guys, what are we paying you for again?). After a lot of hesitation I work up the nerve to tell my tech lead who is SURPRISINGLY understanding about it and he says:

“Just use the Linux undo command.” Tech nerd joke, since there is no undo command.
I decide to do something that is a long shot; remember that we have dual servers? They are pretty much the same. I copy the data from /xyz/data on the other server to the server that has been broken. I cross my fingers and go into the XYZ tool to see if it loads up and all the data is there.

It is. Whew.

In that situation I was lucky, but in many other situations developers aren’t so lucky and one misplaced space could mean significant effort wasted on data recovery problems and competitors gaining an edge. Contact us to learn more about effective backup solutions that will help you avoid disaster.

Dale Shulmistra is a Business Continuity Specialist at Invenio IT, responsible for shaping the company’s technology initiatives -- selecting, designing, implementing & supporting business continuity solutions to bolster client operational efficiencies and eliminate downtime.