My Kingston SA400S37480G 480GB SATA SSD is Disk 0, divided into D: and E:. Windows is installed on a separate C: drive. D: contains some programs and games, while E: mainly contains Steam, Valorant, and HoYoPlay/ZZZ.
Everything worked normally until HoYoPlay stopped responding. After I tried to close it, the launcher went black and E: suddenly began showing 100% disk usage even though read/write speeds were near zero. Response times jumped from normal values to thousands of milliseconds, reaching extremely high levels at times. Anything that accessed E: became very slow or froze, while Windows, C:, and programs installed on D: continued working normally.
Windows was unable to repair E:. Event Viewer showed repeated messages saying that I/O operations for Disk 0 were being retried. CrystalDiskInfo reported the SSD as healthy, with no reallocated sectors or bad blocks, although it showed uncorrectable errors and a high SATA PHY/CRC-related raw value. I am not sure how meaningful those vendor-specific SMART values are.
Even deleting files, emptying the Recycle Bin, scanning the partition, or attempting a Quick Format caused Disk 0 to remain at 100% usage and made the system sluggish. Eventually Windows reported that E: needed to be formatted.
I removed the SSD and connected it to another computer. The partition formatted normally there. After reinstalling the SSD in the laptop, I recreated E:, ran disk checks, checked its health, and installed a program on it. Everything currently works normally.
Could this indicate a failing SSD, a loose or faulty SATA connection, or file-system corruption? Is it possible that the launcher caused the problem, or did it merely expose an existing hardware or connection issue? I am especially confused because D: remained usable even though it shares the same physical SSD, and the drive now appears to be functioning normally again.
4 Answers
It is also possible that the partition or file system became corrupted rather than the NAND itself failing. Removing the partition and recreating it effectively discarded the old file-system metadata, which could explain why the drive became usable again. The other computer may also have handled the SATA connection or formatting process more reliably.
Still, software corruption alone would not fully explain repeated I/O retries and a very high CRC-related count. Those errors point toward either communication problems or a drive that is beginning to fail, so I would not consider the issue resolved just because the new partition currently works.
A problem can affect one partition while the others appear fine. If the damaged files, metadata, or unreadable sectors were concentrated in E:, accesses to that volume could hang while D: continued working. However, the fact that formatting succeeded when the SSD was connected to another computer also makes the laptop's SATA connection, power delivery, controller, or driver worth investigating.
The safest approach is to treat the drive as unreliable until proven otherwise: keep current backups, avoid storing the only copy of important data on it, and run the manufacturer's diagnostic tools. A temporary recovery does not rule out an intermittent hardware fault.
The launcher probably did not damage the SSD. More likely, the drive or its connection started having trouble first, which caused the launcher to freeze while it was accessing E:. A device that repeatedly retries failed I/O can show 100% active time with almost no actual throughput and extremely high response times.
The uncorrectable-error count and the large SATA PHY/CRC value are worth taking seriously, even if the overall health percentage looks good. SMART health percentages are not proof that every part of the drive or its connection is reliable. Back up anything important from the SSD and monitor it closely. If the symptoms return, replace the drive and inspect or replace the SATA cable or connector.
The most useful next step is to test the SSD with a different known-good cable, connector, or enclosure while checking whether the CRC count increases. CRC errors generally indicate data-transfer problems between the drive and host, often caused by a cable, connector, interference, or controller rather than bad sectors alone.
If the count continues rising or the I/O stalls come back, replacing the SSD is the sensible choice. Even if the SSD survives a format and reports a high health percentage, intermittent timeouts are enough reason not to trust it with important files.
That makes sense. The new partition passes the disk check and is working normally for now, but I will keep backups and watch for the errors returning rather than assuming the drive is fixed.