Interpreting S.M.A.R.T. data on Sun F40 (LSI/Seagate WarpDrive)

Markess

Active Member
May 19, 2018
471
190
43
I got the 400GB F40 model of one of these a while back but never got around to plugging it in: https://forums.servethehome.com/ind...e-sun-oracle-f80-800gb-pci-e-ssd-95-fs.20922/

I remembered that I had it when the thread got revived recently, and I decided to test it out. I'm not an expert at reading S.M.A.R.T. values and the internet seemed contradictory, so thought I'd ask here for help interpreting these:

- All four drives on the card come back "healthy" when tested with systemctl in linux. All the attributes that I understand are most critical (05, 181,182,187,188,197,198, etc) are all 0. But....

-- All four drives have raw values in the billions (TRILLIONS for one drive) for attribute 13 "Soft Error Read Rate".

-- One of the four also has the exact same raw value of 4294967296 in attributes 195 "Hardware ECC Recovered", 201 "Soft Read Error Rate", and 204 "Soft ECC Correction". From what could find on the internet, if there's repeated failed read/write attempts that bump the numbers in these attributes, eventually the drive will do some mapping/reallocation. But as noted above, the count for that is 0.

So I don't know what to make of it. The wishful thinking part of me hopes its just an oddity of the product itself, or maybe LSI/Seagate/Sun use those attributes for something else?

If anyone'sgot knowledge of these, any advice would be appreciated!
 
Last edited:

Markess

Active Member
May 19, 2018
471
190
43
I decided to try this drive in my desktop, which dual boots Windows, and installed Crystal Disk Info. That program tells me all 4 drives are healthy and 100%.

Unlike systemctl on linux (which reported "device is not in database), Crystal Disk Info seemed to recognize the disks for what they were. So, maybe Crystal Disk knows enough about these drives to recognize these odd values as non-critical and report that they are healthy?