Last Updated on August 3, 2017 by Dave Farquhar
Longtime reader Dan Bowman–probably my very first reader, come to think of it–sent in this article from Infoworld regarding SSDs and data loss in power failure.
It’s not theoretical. I’ve seen it. I also know how to prevent it.
I lost my first OCZ Vertex to a power failure. I recovered the data using Recovery Is Possible, but lost my filenames and file attributes. I got my data back, but literally only my data. That hurt. So I ended up using a backup and losing a couple of weeks’ worth of files.
I’ve seen hard drives go berserk too when the power goes out too. Journaling filesystems don’t stop them from trashing MBRs for some reason. SSDs may be more prone to failure and catastrophic corruption than traditional hard drives are when that happens, or I may be unlucky.
Some drives put giant capacitors onboard to give the drive enough time to finish whatever it was doing after the host computer loses power. That seems to help.
But the best solution is to keep the power from going out in the first place. Get a UPS, and use it. An APC BE350G costs $40, and gives a computer enough time to shut down gracefully, at the very least. And if you have a datacenter and it has no backup power whatsoever, I know three things about APC rack-mount units: They work well, they weigh a lot, and they’re cheaper than data recovery.
The UPS is an additional cost, but it makes your computer work better and last longer anyway, and it takes the biggest risk out of using an SSD.
I’m not comfortable with the article’s conclusion that SSDs aren’t ready for prime time because they corrupt when you power them down 3,000 times. After all, if you pull the plug on a hard drive 3,000 times the results won’t be pretty either. The observation that SSDs don’t like power failures is nevertheless a known problem. However, dismissing new and beneficial technology due to a weakness isn’t the answer, either. You put in something that compensates for the weakness. When the compensating measure is a $40 UPS, it’s a small price to pay for the other benefits. And if you forget the UPS, here’s how to fix it using the power cycle method.
David Farquhar is a computer security professional, entrepreneur, and author. He started his career as a part-time computer technician in 1994, worked his way up to system administrator by 1997, and has specialized in vulnerability management since 2013. He invests in real estate on the side and his hobbies include O gauge trains, baseball cards, and retro computers and video games. A University of Missouri graduate, he holds CISSP and Security+ certifications. He lives in St. Louis with his family.
I repeat my comment to the source article:
I don’t understand running any system with information important to the user without a good UPS, I prefer extended run time as well.