Why is this SSD Failing?

[linster]

My Agility 3 240 GB is apparently dying. SMART shows 4% life left. But it also shows less than 8 TB host writes. So why is this drive failing? Are there other factors besides write cycles that affect SSD life?

 

[Elixer]

That is just a guestimate anyway, it may continue fine, or it may not.
OCZ is using that to tell you how much NAND life is left.

So, make backups, and keep using it until it dies, or buy a new one.

BTW, I would do a secure erase on it, and see if that does anything.

 

[dave_the_nerd]

Different SSD makers report that data differently. It might not actually be worn out. Or the counter was reset at some point and total host writes is actually a lot higher.

But it is a six-year-old SSD, and of a model and generation that were known for problems and interesting failures.

The error rates looks bad, in any case. So it's presumably eaten through its spare area and is about ready... RIP, little guy.

 

[UsandThem]

This is probably a good read for you, it's a little old at this point, but it addresses the SMART status and other good-to-know things. Rarely do SSDs actually die from too many actual user writes (there have been some longevity tests where they write a lot of data to the drives before they die from that).

The NAND ages from use/heat over time. However, many people replace them well before that or they die for other reasons.

http://www.tomshardware.com/reviews/ssd-reliability-failure-rate,2923.html

Obviously, we know that SSDs still fail, though. All it takes is 10 minutes of flipping through customer reviews on Newegg's listings. But write-cycle exhaustion isn't the problem. Sometimes firmware is to blame. We know this because of the firmware updates vendors issue specifically targeting a documented problem. Other failures are electronic in nature. A capacitor or memory IC might go out, taking the SSD with it.

 

[Glaring_Mistake]

My Agility 3 240 GB is apparently dying. SMART shows 4% life left. But it also shows less than 8 TB host writes. So why is this drive failing? Are there other factors besides write cycles that affect SSD life?

Several of the old OCZ SSDs were prone to failure (according to OCZ this was due to having no choice but to use whatever NAND they could get their hands on) and this is probably just an example of that.

If you look at Program Fail Count and Erase Fail Count (their Raw Values) you can see that a number of times it has failed to write/erase data to/from a NAND cell.
Those failures caused the drive to replace them with some kept in reserve should something like that happen.
Considering the number of reallocated sectors (349) it is probably safe to say that the normalized value for Reallocation Event Count remains at 100 regardless of the real status (something I've seen with other values).
If the normalized value for Reallocation Event Count actually went down as the number of reallocated sectors went up you would likely get to see that the reserves have basically dried up.

 

[linster]

Thanks guys for the responses. My initial reaction was from endurance tests by The Tech Report and the hundreds of TB writes before failure on their drives. I'm seeing way below that on mine. I've never actually learned to read SMART reports. I'm aware of the issues with this model. I've always counted myself lucky that I've never had issues with it. It's actually been a great drive. Started out as my personal drive and then I put it into my work laptop. It has extended the life of the laptop beyond any reasonable lifespan. It's time to ask the boss for a new machine.