Random disk fault, but smartctl is good.

[xk3tchuPx]

Hi,
I do have an issue with a relatively new pool (It's 2 months old).
The pool is a ZFS raid 10 with 4 intel ssdsc2kb240g7.

About every week I get an error on the pool and one device is kicked out because of errors.
I ran smartctl -long and didn't found any error.
Each time it's a different drive, I can't find the root of the issue.

A zpool clear fix the issue for about a week, sometime even less.



The server is running 6.2-6 with 94G of DDR4 ECC memory.


Anybody have an idea of what is going on?

Regards,
xk3tchuPx

 

[Dunuin]

I also had that problem and it was fixed by changing the SATA cable. And it was always the same SSD. Are your drives connected to a PCIe card?

 

[xk3tchuPx]

The server is an echostream flachesan (https://echostreams.com/collections...-processor/products/flachesan1l-d5-1u-20-bays)

I will try to move them to another section, but since it happen with more than one drive, I doubt it's going to help.

R,
xk3tchuPx