Controller health
Almost invisible to SMART. The controller is a small processor running firmware, and when it fails it fails completely — no attribute counts up beforehand. This is the commonest catastrophic SSD failure.
The big one
The attributes are genuinely accurate about flash wear — and flash wear is not what usually kills an SSD. The controller is, and almost nothing in SMART sees it coming.
Percentage used, reserved or reallocated blocks, and uncorrectable errors. Watch those and you have the picture; the several dozen other attributes mostly tell a normal user nothing actionable.
Install CrystalDiskInfo, DriveDx or smartctl — to a different drive — and look for these rather than the overall verdict.
How much of the drive’s rated write endurance has been consumed. It counts up from 0 to 100. At 100 the drive has reached its designed lifetime, which is not the same as failing — many run well beyond it.
The raw figure behind the wear percentage. Useful for sanity-checking: a consumer TLC drive rated for a few hundred terabytes written is nowhere near trouble at 30TB, whatever the percentage suggests.
On NVMe drives. The controller keeps reserve blocks to replace failed ones. Available Spare falling toward the threshold is a genuine warning, and it is the closest an SSD gets to a hard drive’s reallocated sector count.
Uncorrectable errors the controller could not fix with ECC. Any non-zero value on a drive you care about warrants copying data off. Unlike wear, this is damage rather than consumption.
Which is why an SSD can pass every check and be gone the next morning.
Almost invisible to SMART. The controller is a small processor running firmware, and when it fails it fails completely — no attribute counts up beforehand. This is the commonest catastrophic SSD failure.
Composite temperature is reported, but many consumer drives do not log throttling events. An NVMe repeatedly hitting its limit under a graphics card may show nothing unusual.
Most external enclosures do not pass SMART through at all, so the tool reports nothing. That is the bridge, not the drive — and on WD units, removing the disk to test is unwise because the encryption key lives on that board.
Consumer SSDs mostly lack power-loss protection capacitors. An abrupt cut mid-write can leave mapping tables inconsistent, and no attribute predicts that.
The practical consequence for how you should treat each.
With a hard drive, monitoring earns its keep. Reallocated and pending sector counts climb over days or weeks while the drive is still readable, and that window is where the value sits — you get warning, and warning is actionable.
With an SSD, the failure that matters usually arrives without any counter moving. So “I will back it up when it starts playing up” is a strategy that works on one and not the other. On solid-state storage the only reliable protection is a backup that already exists, because there is no deterioration phase to catch it in.
Which is worth knowing before it matters rather than after. An SSD showing 4% wear and perfect health attributes is not evidence of safety — it is evidence that nothing measurable has gone wrong yet.
Wear indicators climbing normally. Not urgent. Percentage Used rising steadily on a drive with years of writes behind it is the drive doing exactly what it was designed to do. Plan a replacement eventually; do not panic.
Available Spare falling, or integrity errors appearing. Act now. Copy your data off, most important first, to a different physical drive. Do not run a deep scan or a repair tool — on an SSD those add writes and wear without diagnosing anything the attributes have not already told you.
And if the drive has already vanished from the BIOS or is reporting an absurd capacity, that is controller territory rather than wear. SSD and NVMe recovery in Manchester is from £300 +VAT after a free 48-hour diagnostic.
Install CrystalDiskInfo, DriveDx or smartctl to a different drive and look at Percentage Used, Available Spare against its threshold, and Media and Data Integrity Errors. Ignore the overall Good/Caution verdict, which is a manufacturer threshold.
It counts up from 0 as the drive consumes its rated write endurance. Steady increase over years is the drive working as designed, and many run well beyond 100%. It is a wear counter, not a failure prediction.
Because SMART reports flash wear accurately and controller health barely at all — and the controller is what usually fails catastrophically. No attribute counts up beforehand, so a drive can pass every check and be gone the next morning.
The reserve blocks an NVMe controller keeps to replace failed ones. It falling toward its threshold is a genuine warning — the closest an SSD gets to a hard drive’s reallocated sector count, and worth acting on.
No, it usually means the USB bridge does not pass SMART through. That is the enclosure rather than the drive. Be careful about removing the disk to test it, particularly on WD units where the encryption key lives on the bridge board.
No. Deep scans and repair utilities add writes and wear without diagnosing anything the attributes have not already told you. If the numbers are worsening, copy your data off instead.