My one-year-old 1TB ADATA XPG SX8200 Pro is connected as an external drive through an ORICO M2PV-C3 enclosure using a Realtek RTL9210B controller. It worked normally for months, but its SMART data suddenly began changing in August. Available Spare gradually fell from 100% to 0%, while Critical Warning changed to 9 and Media and Data Integrity Errors rose from 0 to 247. After downloading a 3GB file, the integrity-error count jumped to 9,245. CrystalDiskInfo still reports 100% health based on the drive's relatively low 3,125GB of host writes, but marks the drive as Bad in red. The SSD remains responsive, files can be copied from it, and large files have matched SHA-256 checksums after being copied elsewhere. It has not obviously become read-only or developed noticeable slowdowns. The laptop and enclosure usually remain powered 24/7, and CrystalDiskInfo polls the drive every five minutes to prevent the enclosure from sleeping. What do these conflicting SMART values actually mean? Could this be a firmware or enclosure-reporting bug, or is the SSD failing despite still appearing to work normally? Should I stop using it, replace the enclosure, or pursue a warranty claim?
4 Answers
The two health indicators measure different things. The 100% figure mainly reflects endurance consumed, and 3,125GB written to a 1TB drive is only a few full-drive writes, so the NAND may show almost no normal wear. Available Spare is different: it represents replacement blocks reserved for failures. Dropping from 100% to zero means the controller has been retiring failing blocks and has exhausted that reserve. A drive can therefore report 100% remaining life while also being physically unhealthy.
I would not keep using this as a working or gaming drive. First make at least one verified backup, then test it directly in an M.2 slot if possible and inspect the manufacturer's diagnostic tools or firmware. Keeping an SSD in an enclosure does not automatically kill it, and only a small number of unsafe shutdowns does not rule out a controller, NAND, power, or firmware fault. Even if the enclosure is responsible, the current SMART state is unsafe to ignore. If the serial number and purchase documentation are available, a warranty claim is worth considering; a drive reporting zero spare blocks and critical warnings is defective regardless of its percentage-health estimate.
A rapidly increasing Media and Data Integrity Error count is the most concerning part, especially the jump from 247 to 9,245 after one 3GB write. Successful playback or copying does not prove that every sector is healthy, although matching checksums on copied files is a useful sign for those specific files. Stop adding data, copy everything to another verified destination, and compare checksums instead of relying only on whether files open normally.
The enclosure could also be involved. A USB bridge can introduce link resets, power-management problems, or misleading SMART translations, so testing the bare M.2 drive in a motherboard slot could help separate an enclosure issue from an SSD failure. However, zero available spare and a critical warning should still be treated as a replacement-level problem.
Critical Warning is a bit field rather than a simple error count. A value of 9 indicates multiple warning bits are set, including the available-spare threshold and a warning associated with read-only operation. Some controllers report that state before they fully block writes, or only enforce it after another reset. The fact that new files can still be written does not make the drive safe; it may simply not have reached the point where the failure becomes obvious to Windows.
That would explain why this drive behaves differently from another SSD that became obviously read-only and froze when accessed. The current one may still be readable because it has not completely failed yet, rather than because the warnings are harmless.
Large files copied from the SSD have matched SHA-256 checksums, so the data I tested appears intact. Could the SMART values still be a firmware reporting bug, or does the zero spare count remain enough reason to treat the drive as failed?