Maxim_Kazmin - Fotolia
5 biggest data backup mistakes -- and how to avoid them
Don't let a backup blunder hurt your organization or career. Here are five common, yet potentially disastrous data backup mistakes you must avoid at all costs.
Everybody makes mistakes, but when a slip-up involves data backups, the consequences can be catastrophic to both the enterprise and the mistake-maker's job security.
Don't let a data backup error ruin your day -- or career. Here's a rundown of the five biggest data backup mistakes you need to avoid.
1. Failing to store backup data at a separate physical site
Stockpiling backups at your primary business site means you might be unable to recover them when the need arises, since they will likely be exposed to the same danger as your primary files. "Off-site backups are a good way of ensuring the ability to recover," said Troy Wilkinson, CEO of Axiom Cyber Solutions, a cybersecurity management firm.
Remember to physically secure backups, or your priceless recovery data could suddenly vanish due to theft, negligence or the actions of a disgruntled employee. You might not discover that the backups are missing until it's too late.
2. Not auditing backups for freshness and integrity
Many organizations simply install and configure backup programs and then check off the task as if everything has been completed, said Darren Gallop, CEO of cybersecurity and privacy compliance firm Securicy. Yet, over months and years, IT staff change file structures, create new folders and add new applications to the tech stack. Then, when they're needed most, the old backup scripts are discovered to be obsolete and effectively useless. "Without conducting periodic backup audits, this oversight can go unnoticed for a long time," Gallop said.
Don't make these data backup mistakes. Even small enterprises with limited resources should comprehensively test their backups every three months, recommended Brian Gill, CEO of Gillware, a data recovery firm. "Don't forget about proprietary software, software licenses and installers," he said. "You need to actually mimic the real-world situation of the disaster and prove that the backup is comprehensive and complete."
There's no need for expensive tech or consultants to run your own backup audits, Gallop observed. "You just need to develop a procedure and determine the appropriate cadence for the audits."
3. Believing RAID is its own backup system
Many different types of organizations rely on a redundant array of independent disks (RAID). These systems distribute data and parity across a set of disks, protecting files stored on the array against the failure of any individual disk.
Unfortunately, even these highly redundant systems are not completely infallible, noted Mike Cobb, director of engineering for data recovery firm DriveSavers. Initial failures sometimes go undetected until a secondary failure has occurred. "A RAID 5 configuration, for example, can only offset the failure of a single disk drive," Cobb said. "However, cases of multiple drive failures happen more frequently than one might imagine."
It's important to remember RAID systems are fault-tolerant, not fault-proof. "A big mistake is that businesses -- particularly SMBs -- do not back up their RAID systems, thinking that RAID is its own backup," Cobb said. "We see this mistake made with all RAID configurations but most often with RAID 0."
4. Allowing protected servers to access the backup system
"When the backup is online, available and able to be deleted, calamity is just a keystroke away," said Adam Stern, CEO of cloud service provider Infinitely Virtual. In the days when tape was king, backed-up data was physically segregated and placed safely out of harm's way. Now that online backup rules, backups need to be set to read-only.
Rely exclusively on SAN snapshot-based backups -- backups controlled by the SAN, Stern advised. Storage experts consider this read-only technique to be the most reliable backup methodology. It's also generally regarded as the easiest process to manage. "In theory, users can back up thousands of servers with just a few jobs," Stern noted.
Remember, too, that, when backup storage is online, it can be compromised by ransomware that's designed to traverse the network. "It's important that the backups are segmented and, if possible, have separate credentials [other] than standard administrator accounts to avoid possible [malicious] encryption of the backup files," Wilkinson added.
5. Depending on only a single backup approach
If you use only one storage medium and then store that data in a single location, that opens the door to disaster. "Data is the lifeblood of most organizations, and by not dispersing the data across a multitude of geographies and storage mediums with at least one air gap, organizations risk data loss from hackers, viruses, human error and natural disasters," said David Feller, vice president of product management and solutions engineering at data storage vendor Spectra Logic.
An organization can avoid this type of the common data backup mistakes by following the 3-2-1-1 rule. "Critical data should be safeguarded with both physical and genetic diversity," Feller noted. The 3-2-1-1 rule requires maintaining three copies of business data, stored on two different types of media, with one copy stored on-site and one copy stored off-site, with one of those copies physically removed from the network, creating an air gap that cannot be hacked or ransomed.
