My external 1TB ADATA XPG SX8200 Pro had flawless SMART readings for about a year, with only around 3.1TB written and temperatures around 30–40°C. After my laptop froze while I was playing a game and browsing, CrystalDiskInfo suddenly reported 9 critical warnings, 0% available spare, and 247 media and data integrity errors. The available-spare value had already been gradually falling from 100% to 0% over the previous couple of weeks, despite very little writing. Later, writing a 3GB file caused the reported integrity-error count to jump to 9,245, even though the download completed at normal speed and the file remained readable after reconnecting the drive. The SSD is installed in an ORICO USB enclosure using a Realtek RTL9210B controller and is kept connected continuously. It still appears fast and usable, so I'm trying to understand whether the drive is actually failing, whether the enclosure or SMART translation is producing bad data, and why it has not switched to read-only mode.
1 Answer
Treat the drive as failing even if it still appears completely normal. Critical warnings, zero available spare, and rapidly increasing media or data-integrity errors are not values to ignore; SSDs can continue serving reads and accepting writes until the controller or flash fails more dramatically. Copy anything important to another disk immediately, then stop using this one for games or further testing. After your data is safe, reseat the SSD and, if possible, test it in another compatible enclosure or system to rule out a USB bridge, cable, power, or SMART-reporting problem. If the same warnings appear, replace or warranty the SSD. The fact that it has not entered read-only mode does not mean it is healthy or that future writes are safe.
I understand the warnings are serious, but there were no slowdowns or corrupted files, and even the large write completed normally. Could the USB enclosure be misreporting the NVMe SMART fields rather than the NAND actually failing?