My boot SSD suddenly showed 2% health in HDD Sentinel, but I didn't replace or reformat it because I was putting off reinstalling Windows and my applications. Later, the computer crashed with a BSOD and wouldn't boot. After basic troubleshooting, I replaced the SATA cables and the system started working again. When I checked the SSD afterward, its reported health had returned to 100%. I've restarted the computer several times and the reading is still 100%, so I'm wondering whether the cable caused a misleading health reading or whether the SSD is actually failing. Should I trust the current result, check specific SMART values, or replace the drive anyway?
3 Answers
A bad SATA cable can cause communication errors, corrupted transfers, and increasing SATA CRC error counts. Some drive-monitoring programs may factor those errors into their own health estimate, so replacing the cable can make the reported percentage look normal again. However, a cable cannot repair worn-out flash memory, so you should still investigate the underlying SMART data and keep a current backup.
HDD Sentinel’s health percentage is based on its interpretation of the drive’s SMART attributes; it isn’t necessarily a universal SSD-health measurement. Without seeing the SMART attribute list, it’s impossible to know whether the 2% rating came from CRC errors, media wear, uncorrectable errors, or another warning. Check the drive with another SMART utility as well and look for reallocated or uncorrectable sectors, media errors, wear indicators, and interface CRC errors.
Even if the cable was the immediate cause of the boot problem, a drive that once reported 2% health should be treated cautiously. Back up anything important now, avoid relying on the SSD as the only copy of your data, and consider replacing it if SMART shows actual media or wear failures. Also check the replacement cable and motherboard SATA port if CRC errors continue increasing.
So the 100% reading may not prove that the SSD is completely healthy? I’ll check the individual SMART values instead of relying only on the percentage.