The Drive Health Check reads a drive's own SMART self-monitoring data — the statistics every hard drive and SSD already keeps about itself — and turns it into a plain report: power-on hours, temperature, defect counts, and a single health score you can act on. It is read-only: nothing is written to, repaired, or reset on the drive. This walks through running the check, reading what it reports, and deciding what to do with the result.
Before you start
There are two ways to run the check, and which one to use depends on the machine in front of you.
- Bootable toolkit. Boot the machine into the free ITAD Tools iPXE toolkit and choose Drive Health Check from the menu. No account is needed to run the check. Submitting the result with an organisation key is optional, and is what enables per-serial history in the Drive Health Viewer — this suits a machine you can take offline for a few minutes.
- Host agent. Sign in, open Boot Tools, and download the Windows or Linux build. Use this on a machine that needs to keep running — for example, one that is already partway through a wipe — because it reads drives on the operating system that is already booted. It requires a free account that belongs to an organisation, since it submits results using your organisation key — signing up on its own does not create one; create or join an organisation after you sign in.
Either way, nothing has to come out of the machine: the check reads the drives in place.
Reading a drive step by step
- Pick a path from "Before you start" above, based on whether the machine can go offline or needs to keep running.
- Bootable toolkit: boot the machine from the ITAD Tools iPXE image and choose Drive Health Check from the menu when it appears.
- Host agent: set your organisation key in the environment, then run the agent with administrator privileges on Windows or as root on Linux — reading raw disks needs elevation.
- The check reads each connected drive's SMART data directly from the drive itself. It only reads; it never writes to, repairs, resets, or remaps anything.
- Bootable toolkit: when the check finishes, a verification code appears on screen. That's as far as the free, no-account path goes. To keep the result, sign in, open the Drive Health Viewer, and enter the code to attach the report to your account.
- Host agent: because each submission already carries your organisation key, the server links it to your organisation automatically — no code to claim afterward.
- Sign in and open the Drive Health Viewer to read the full report for every drive the check found.
Reading the results
Once a report reaches the Drive Health Viewer — automatically if you used an organisation key, or after you sign in and enter your verification code — it shows, for each drive:
- Health status — the PASSED / WARNING / FAILED status derived from the drive's own SMART self-assessment.
- Health score — a single figure combining the drive's wear, endurance, and defect indicators, so a worn SSD or a drive collecting bad sectors stands out immediately.
- Power-on hours and power cycle count — how long the drive has run, and how many times it has started and stopped.
- Reallocated, pending, and uncorrectable sector counts — the values that show real media defects (see "Making the call" below).
- Wear level and total bytes written — for SSDs, how much of the drive's rated endurance has been used.
- Temperature and self-test status — the drive's current temperature and the result of its own built-in self test, where the drive supports one.
Making the call
None of these values mean anything by themselves — how you read them depends on what you plan to do with the drive. A few conventions are close to universal:
- Uncorrectable sectors. Any non-zero count means the drive already failed to recover some of its own data. Most buyers and warranty programs treat this as an automatic reject, not a borderline case.
- Pending sectors. A pending sector can still resolve itself if a later write succeeds. Read the drive again before deciding — a pending sector that survives a second check is the one to treat as failing.
- Reallocated sectors. This count never decreases, so a drive with some history of reallocations is common and often still sellable. What matters is direction, not the reading itself: a count that holds steady across repeat checks is widely accepted on used media; a count that keeps climbing is not.
Age, wear, and temperature work the same way — none of them has a single cutoff that applies to every buyer, warranty term, or drive class. Set your own lines to match who you sell to, and read a drive more than once before letting a defect count decide its fate.
When something looks wrong
- Windows flags the host agent as an unknown publisher. The build is unsigned, so this warning is expected. Verify the published checksum before you run it, then continue past the prompt.
- The host agent will not read any drives. Reading raw disks needs elevation — run it from an Administrator PowerShell window on Windows, or with root on Linux.
- A drive shows no SMART data. Some USB enclosures and older RAID controllers do not pass the drive's own reporting through. Connect the drive directly, or read it with the bootable toolkit instead.
- You closed the bootable toolkit without noting the verification code. The check is read-only, so there is no harm in booting it again to get a fresh code.
Frequently asked questions
Does checking drive health change anything on the drive?
No. The check is read-only. It reports the statistics the drive already keeps about itself and never writes, repairs, remaps, or clears a counter.
Do I need an account?
The bootable toolkit is free and needs no account. The host agent, which runs on a machine that is already switched on, requires a free account that belongs to an organisation, since it submits results using your organisation key. Signing up on its own does not create one -- create or join an organisation after you sign in.
Can I check drives while a wipe is running?
Yes. The host agent reads SMART data on the running operating system, so the drives stay where they are and the wipe keeps going.
Does it work on SSDs as well as spinning disks?
Yes, though the values that matter differ. Endurance and total data written describe an SSD better than age does, while spinning disks are judged mostly on media defects.
What if a drive reports no SMART data at all?
Some USB bridges and older RAID controllers hide the drive's own reporting. Connecting the drive directly, or booting the toolkit instead, usually exposes it.
Try it
Bootable toolkit. Free, no account needed to run the check.
Host agent. Reads the same SMART values on a machine that is already running. Requires a free account with an organisation.