This is the 283rd article in the Spotlight on IT series. If you'd be interested in writing an article on the subject of backup, security, storage, virtualization, mobile, networking, wireless, cloud and SaaS, or MSPs for the series PM Eric to get started.
Not long ago I had a situation where something happened to an XML file that I needed. I discovered it was corrupted, and I needed to do a quick restore.
“No worries,” I thought, “I’ll simply copy the most recent version that I had saved to my file server.” No such luck. That file had been just copied and was damaged.
“OK, no worries,” I thought. “That’s why I have a periodic backup copy.” It turns out that had been working flawlessly. Except there was a catch — it had been backing up the damaged file. This meant that any and all other copies of the file were also damaged as far back as to when the problem occurred.
3-2-1
There is an old expression, mantra, rule or recommendation pertaining to data protection and backup of “3-2-1.” This means keeping at least three copies (or more) of anything important, placing those copies on at least two different mediums or devices, and have at least one of those copies off-site away from your system.
But, this simple rule isn’t prefect. What the 3-2-1 rule doesn’t explicitly mention is that there should also be unique copies or versions that are offset by some time interval to reflect that state or context of an application or your data as of when it was protected.
What this means is that you may have three or more copies on two or more devices (SSD, USB, HDD, SSHD, tape) including one or more offsite (on-line at a cloud service provider or alternate site, or off-line). However, what if all of those copies are the same? Worse yet, what if those copies also contain the file, document, object, database, photo, video or other entity that you need, and it is corrupted or otherwise useless?
My corrupted file
The good news was that I did have different versions of the files, and more than three copies on more than two devices (SSD, HHDD, HDD) and at more than one location (local, off-site/off-line on removable HDD, and off-site at my cloud provider, and one other).
The challenge then turned out to simply figure out at what point in time the file was damaged, to then go grab that version and bring it up to date using the contents of another good file.
Granted, for data that is not changing or static, all of those copies may not be needed. However, make sure your main copy is good, and that there are at least three copies, two of which are on different mediums or devices, and in different places. Still, my most important data, files and applications have more frequent copies made to reflect different points in time.
By the way, while cloud is useful, don’t underestimate how much data can fit into a small safe deposit box using 2.5-inch high-capacity HDDs for those master golden copies that are there just in case all else fails or takes to long to bring back from the cloud.
Moral of the story
Test and periodically audit your data protection and backups to make sure not only that they’re running, but that they’re also picking up the right versions of data. This also means checking to see if you can restore (to an alternate location) your data, as well as apply security rights, decrypt or other actions to make it useful.
--
Questions? Comments? Backup tips or tales of your own? Chime in in the comments below!