pool degraded smartd[7132] error

[cpg0000]

I'm getting the following error and can't seem to track down what it means. Any help would be appreciated.

smartd[7132]: Device: /dev/ada5, 1 Currently unreadable (pending) sectors

The drive is part of an 8 drive raidz2 array with each drive being 4 TB drive and the entire system is about 2.5 years old. The drives are 4 Seagate, 3 HGST, and 1 WD. ADA5 is the single western digital drive and the newest drive in the array. It was replaced after the first drive also assigned ADA5 went bad after a year and became unreadable.
When viewing the reports ADA5 always has significantly different Disk_Time than the other drives for writes. Not sure what it is trying to do.

version and other build detail if that helps
FreeNAS-11.2-U5
16gb ECC ram
Intel(R) Atom(TM) CPU C2750
AsRock Rack C2750D4I

 

[cpg0000]

Bump
Any ideas?

 

[AlexGG]

"Current pending sectors count" is a SMART measurement basically indicating number of defective sectors on the disk surface. You have 1 now. There are several kinds of these (pending, reallocated, offline uncorrectable), and different people have different levels of tolerance (should I bin the drive which only has one? one of each? ten maybe?). However, general recommendation is to replace the drive. What you do with the drive then, there are some possibilities, run a full SMART test on it, try to return under warranty if possible (if SMART test uncovers more errors and declares it faulty), zero fill it and see what happens, relegate it to temporary storage, just bin the damn thing, whatever. In practice, drive which developed one defect is likely to develop another soon. Not always, but likely.

 

[cpg0000]

Thanks Alex. That confirms my suspicion and what I was thinking I should do. I found it strange that 2 drives from 2 manufacturers in a row died in the same spot in the pool after about a 1-1.5 years each.
I’ll definitely start shopping for a replacement

"Current pending sectors count" is a SMART measurement basically indicating number of defective sectors on the disk surface. You have 1 now. There are several kinds of these (pending, reallocated, offline uncorrectable), and different people have different levels of tolerance (should I bin the drive which only has one? one of each? ten maybe?). However, general recommendation is to replace the drive. What you do with the drive then, there are some possibilities, run a full SMART test on it, try to return under warranty if possible (if SMART test uncovers more errors and declares it faulty), zero fill it and see what happens, relegate it to temporary storage, just bin the damn thing, whatever. In practice, drive which developed one defect is likely to develop another soon. Not always, but likely.

 

[Jailer]

Thanks Alex. That confirms my suspicion and what I was thinking I should do. I found it strange that 2 drives from 2 manufacturers in a row died in the same spot in the pool after about a 1-1.5 years each.
I’ll definitely start shopping for a replacement

With only 1 pending sector you could keep an eye on it to see if it gets worse. You should also have a spare drive burned in and ready to go for just such an occasion.

With 2 drives from the same location I'd check the drive temperatures and monitor them closely. If you have a drive that's been getting too hot it will shorten it's life span significantly.