How it works

SMART (Self-Monitoring, Analysis and Reporting Technology) is firmware inside nearly every modern HDD and SSD that tracks internal health counters and exposes them to the host over ATA, SCSI, NVMe, or USB–SATA bridges that pass the commands through. Classic ATA SMART attributes are numbered values—Reallocated_Sector_Ct, Current_Pending_Sector, Offline_Uncorrectable, Spin_Retry_Count, Temperature_Celsius, Power_On_Hours—each with a raw count and a normalized value the drive vendor defines. The drive firmware marks a sector as pending when a read fails, then remaps it to a spare if a later write succeeds; the reallocated-sector count climbs when that spare is consumed. On NVMe the same idea appears as SMART/Health Information log pages (media errors, available spare, percentage used) instead of the old ATA ID table.

Host tools read those counters without touching user data: smartctl from smartmontools on Linux/macOS/Windows, CrystalDiskInfo on Windows, drive bay vendor utilities, and NAS packages such as Synology SMART tests or Seagate IronWolf Health Management. A short SMART self-test exercises a limited surface check in minutes; an extended/long test walks the whole media and can take many hours on a multi-terabyte HDD. USB docks and enclosures only help if the bridge chip supports SMART pass-through (often advertised alongside UASP)—otherwise smartctl sees a generic USB device and no attributes.

When it matters

It matters the moment a new drive arrives, during burn-in before you trust a RAID or backup set, and any time a NAS, desktop, or dock starts throwing read errors. Rising reallocated or pending sectors after a full surface write/read pass is the classic infant-mortality red flag: return the drive while warranty is clean rather than rebuild an array around it. Temperature and power-on hours are weaker signals alone, but a drive that runs hot in a packed bay or shows sudden raw-read-error spikes deserves a long self-test and a backup check.

Shopping and workflow traps: buying SMR NAS drives that look fine at idle then thrash during RAID rebuilds (prefer CMR NAS HDDs), using a BOT-only USB dock that hides SMART, and trusting a single “healthy” dashboard color without reading raw reallocated/pending counts. Pair a CMR NAS HDD with a UASP + SMART dock when you burn in bare drives on a PC before they go into the chassis—see How to Burn-In Test a New Hard Drive.

A CMR NAS HDD you can SMART-test during burn-in, a second NAS CMR option with vendor health hooks, and a USB-C SATA dock that exposes SMART pass-through for bench testing.

The WD Red Plus 4TB (WD40EFPX) is a CMR NAS HDD—the class of drive you want to run SMART short/long tests and a full burn-in surface pass on before it joins a home NAS.

The Seagate IronWolf 4TB (ST4000VN006) is another CMR NAS drive with IronWolf Health Management on popular NAS brands, useful when you want vendor health warnings on top of raw SMART.

The ORICO USB-C SATA Dock (DD18C3) is a tool-free 2.5/3.5-inch dock with SMART pass-through and UASP so smartctl can read attributes while a bare drive burns in on the bench.

Walk through the full workflow in How to Burn-In Test a New Hard Drive. For recording tech that affects rebuild behavior, see CMR and SMR. For docks that keep SMART and command queuing intact over USB, read UASP. Comparing the two NAS HDDs above? See WD Red Plus vs Seagate IronWolf 4TB.