Synology RAID-5 NAS Has Power Surge

This failed NAS device came to us from a nearby small business located in Fond du Lac. Their Synology DiskStation DS 1512+ NAS device had failed after a power surge. Inside their DiskStation was a 5-drive RAID-5 array. This NAS device was the only backup of their files they had. Without it, all they could do was sit on their hands. Nothing is more frustrating for an owner or the employees of a small business. Fortunately, our power surge data recovery experts were on the case.

Data Recovery Case Study: Power Surge in Synology NAS Device
RAID Level: 5
Total Capacity: 12 TB
NAS Model: Synology DiskStation DS 1512+
Operating System: Windows
Situation: Two drives in NAS device RAID-5 failed after a power surge
Type of Data Recovered: Business-related graphics and photos
Binary Read: 100%
Gillware Data Recovery Case Rating:

NAS devices are useful tools for small businesses. “NAS” stands for “network-attached storage”. Since these devices plug into a wireless router via ethernet cable, anybody with network access can use the NAS device as if it were attached directly to their computer. A single NAS device can provide terabytes of storage space for many people.

This graphic design business used the RAID-5 array they’d set up on their NAS device to store graphic design files and photos. RAID-5 is a popular way to arrange a set of hard drives to act as a single storage device. Using special parity data, RAID-5 provides fault tolerance in case one of the drives in the array fails. However, if more than one drive in the array fails, the RAID-5 crashes.

Power Surge Data Recovery

A power surge can cause hard drive failure in many ways. One of the most common ways is that the PCB on the back of the drive can get burned out. A sudden spike or loss of power can also cause the hard drive’s mechanical components to become damaged.

Our engineers inspected the hard drives pulled from the Synology NAS device. Of the two drives that had failed, one had suffered severe damage to its platters. The power surge had caused the drive’s magnetic read/write heads to accidentally touch the surfaces of the spinning platters. The read/write heads gouged out large swaths of the magnetic coating on the platters. We refer to this type of platter damage as rotational scoring.

The magnetic coating on each platter of a hard drive contains its data. If any of that coating is scraped off, the data it stored cannot be restored. All rotational scoring results in some irretrievable data loss. Severe enough rotational scoring renders a hard drive unsalvageable.

Rebuilding the RAID-5

Fortunately, a RAID-5 array can function minus one hard drive due to its parity data. Our data recovery engineers had to get as good of a forensic image of the other failed drive as possible.
The other failed hard drive ran rough at first, but our cleanroom data recovery engineers were eventually able to get it into good shape. With full drive images of four of the five hard drives, the case was handed off to our RAID engineer Cody.

When Cody gets his hands on a RAID case, the first thing he does is check the metadata on each hard drive. Putting a RAID back together after it’s been in the cleanroom is a puzzle. The RAID controller inside any NAS device knows how to solve this puzzle. By reading the metadata, the controller knows how the hard drives have to be arranged. If the hard drives didn’t have all this metadata, everything on them would be indecipherable. RAID-5 chops up all of the data the user creates into little stripes. These stripes are typically around 64 kilobytes in size and are spread across all the hard drives.

If the hard drives in a RAID-5 aren’t put together in the right order, anything larger than the stripe size gets rearranged. Since many people have important files larger than 64 kilobytes, getting the order right is very important.

Cody didn’t have the RAID controller for this NAS device. But for this power surge data recovery case, he didn’t need it. As is the case with our other RAID data recovery cases, all Cody needed was the metadata itself. Cody knew how to read this metadata just like a RAID controller does.

The results of this NAS data recovery case were excellent. One hard drive had been rendered completely useless by severe rotational scoring. But a full binary read on the other failed drive gave us all we needed to rebuild the RAID. This was a complete and full recovery.

A sudden power spike or power loss can have unpredictable effects on hard drives. The more hard drives you have, the more of a chance you have of one or more of them failing. In this power surge data recovery scenario, the NAS device’s single-drive fault tolerance wasn’t enough. But fortunately, our data recovery team was able to swiftly reunite this small business with its critical data.

Will Ascenzo
Will Ascenzo

Will is the lead blogger, copywriter, and copy editor for Gillware Data Recovery and Digital Forensics, and a staunch advocate against the abuse of innocent semicolons.

Articles: 213