To test whether a hard drive or SSD is failing on Windows, use Seagate SeaTools for active read/write error tests and CrystalDiskInfo to read the drive’s SMART data. Install SeaTools from seagate.com, pick your drive, run a Quick Test (takes 2 minutes), and if that passes, run an Extended Test (takes hours, reads every sector). CrystalDiskInfo reads SMART attributes directly — a Caution or Bad status, any Reallocated Sectors Count above zero, or a Pending Sector Count means the drive is failing and needs replacing.
Applies to: Windows 10 (22H2) and Windows 11 (23H2, 24H2, 25H2) | Last updated: April 15, 2026
Key Takeaways
- SMART data (via CrystalDiskInfo) reports what the drive itself has recorded — reallocated sectors, pending sectors, and uncorrectable errors. Any non-zero values here mean the drive is already losing data.
- SeaTools Quick Test takes about 2 minutes and catches obvious failures. Extended Test reads every sector on the drive and takes 4–12 hours — run it overnight if Quick passed but you still suspect the drive.
- A Good status in CrystalDiskInfo with a Quick Test pass means the drive is healthy. A Caution or Bad status means back up data immediately and plan a replacement — the drive is failing.
- If a USB-connected drive shows intermittent SeaTools errors but CrystalDiskInfo shows Good, the fault is probably the enclosure or cable, not the drive. Test the drive directly over SATA before replacing.
- If Windows won’t boot, run SeaTools from a Ventoy rescue USB or boot into Hiren’s BootCD PE — both can test the drive without a working OS install.
Quick Steps
- Download and install Seagate SeaTools for Windows.
- Download and run CrystalDiskInfo (portable version works without install).
- In CrystalDiskInfo, check the Health Status and SMART attributes for every drive.
- In SeaTools, select the drive and click Quick Test.
- If Quick Test passes but you still suspect the drive, run Extended Test (several hours).
- If either tool reports errors, back up your data and plan a drive replacement.
Why Two Tools, Not One
SeaTools and CrystalDiskInfo test different things — and a failing drive can pass one and fail the other, so I always run both. SeaTools actively reads from the drive, looking for sectors that won’t respond or return corrupted data. CrystalDiskInfo reads the SMART (Self-Monitoring, Analysis, and Reporting Technology) counters that the drive’s firmware has been quietly tracking since first use. A drive might pass an active read test because the bad sectors haven’t been touched recently, but SMART will already show the fault.
In my repair shop, CrystalDiskInfo is the first thing I run on any drive — if SMART shows Caution, I don’t even bother with SeaTools. I tell the customer the drive needs replacing and focus on rescuing their data. SeaTools comes out when SMART looks fine but the customer is seeing symptoms.
Tip: Most drive manufacturers also publish their own diagnostic tools (Samsung Magician, WD Dashboard, Crucial Storage Executive, Kingston SSD Manager). Those are worth running as a third opinion — they sometimes expose manufacturer-specific SMART attributes that the generic tools don’t interpret correctly.
Step 1: Check SMART Data with CrystalDiskInfo
Start here. CrystalDiskInfo reads SMART data — the drive’s own internal health log — and interprets it with a colour-coded Health Status for every connected drive.
- Go to the CrystalDiskInfo download page and grab the standard edition. The portable ZIP works without install if you prefer not to leave registry entries behind.
- Launch
DiskInfo64.exe. - The window shows every drive as a tab at the top. For each drive, look at the Health Status in the top-left.
How to Read the Health Status
- Good (blue): The drive is healthy. No action needed.
- Caution (yellow): One or more SMART attributes have crossed warning thresholds. Back up data immediately and plan a replacement. The drive can fail at any time from here.
- Bad (red): SMART has flagged the drive as imminently failing. Do not use this drive for anything you care about. Clone the data off and replace.
- Unknown: CrystalDiskInfo can’t read SMART from this drive — usually happens with USB enclosures that don’t pass SMART through. Connect the drive directly over SATA or NVMe for real readings.
Critical SMART Attributes to Watch
Under the Health Status, CrystalDiskInfo lists every SMART attribute with its raw value. These are the ones that matter most:
- 05 Reallocated Sectors Count: any non-zero value = bad sectors have been remapped. One or two is early warning; dozens means the drive is actively failing.
- C5 Current Pending Sector Count: sectors the drive suspects are bad but hasn’t remapped yet. Non-zero = trouble.
- C6 Offline Uncorrectable Sector Count: sectors that cannot be read or written at all. Non-zero = drive has lost data.
- C7 UDMA CRC Error Count: data transfer errors over the SATA cable. Usually a bad cable or port, not the drive.
- 09 Power-On Hours: total runtime. Useful context for how much life the drive has left — most consumer drives last 30,000+ hours.
Warning: A Caution status on a drive that holds your only copy of important data is an emergency. Back up everything before running any further tests — stress testing a failing drive can push it over the edge. For drive temperature concerns, see my guide on how to check hard drive and SSD temperature.
Step 2: Run an Active Test with SeaTools
SeaTools complements CrystalDiskInfo by actively exercising the drive. SMART might report fine if recent writes have all landed on good sectors, but an Extended Test will read every block and find sectors that don’t respond.
Download and Install SeaTools
Go to the Seagate SeaTools downloads page and grab SeaTools for Windows. Despite the Seagate branding, SeaTools tests drives from every manufacturer — Western Digital, Samsung, Crucial, SanDisk, Kingston, any of them.
Run the installer with defaults. Launch SeaTools from the Start menu.
Accept the license agreement on first launch.
Select the Drive
The main window lists every physical drive on the system. Tick the drive you want to test. If you’re not sure which is which, check the model name and capacity against what you know about your hardware.
Run a Quick Test First
Click Basic Tests → Short Drive Self Test (also labelled Quick Test in older versions). This runs the drive’s built-in diagnostic routine and takes about 2 minutes. The drive’s own firmware handles the test — SeaTools just reads back the result.
A Pass result doesn’t prove the drive is perfect — it just means the firmware’s short diagnostic is happy. A Fail result is definitive: the drive has told you itself that it’s faulty.
Run an Extended Test If Needed
If Quick Test passed but the drive is still showing symptoms (random freezes, files that sometimes won’t open, long load times), run Basic Tests → Long Drive Self Test (Extended). This reads every sector on the drive and takes:
- SSD: 10–30 minutes depending on capacity
- HDD (mechanical): 4–12 hours depending on capacity
I run Extended Tests overnight on mechanical drives. The PC can do other work during the test — just don’t put it to sleep.
Testing External USB Drives
USB enclosures complicate diagnostics. SeaTools can test drives over USB, but the results are less reliable than a direct SATA connection. I have seen SeaTools report a drive as healthy when connected via USB, only for the Extended Test to abort halfway through — that pattern almost always means the USB enclosure, cable, or bridge chip is faulty, not the drive itself.
If a USB-connected drive gives suspicious results:
- Pull the drive out of the enclosure and connect it directly to a SATA port in a desktop, or use a different known-good USB-to-SATA adapter.
- Re-test with both SeaTools and CrystalDiskInfo.
- If the drive is now healthy, the enclosure was the problem.
When SeaTools Offers to Fix Errors
If SeaTools finds errors, it offers a Fix All button that remaps the bad sectors. Resist the urge to use this as a repair. Fix All tells the drive firmware to stop using those sectors and redirect to spare sectors — that only delays the inevitable. A drive that has started throwing errors has typically exhausted its margin, and the next bad sector is right around the corner.
Important: Treat Fix All as a temporary rescue that gives you time to back up data — not as a real repair. A drive reporting any errors should be replaced, full stop. Modern drives are cheap; data is not.
Testing a Non-Booting PC
If the PC won’t boot into Windows, you can’t run the Windows version of SeaTools. Two options:
- SeaTools Bootable: SeaTools can create a bootable USB with a Linux-based test environment. Useful for an offline test on a machine that otherwise refuses to POST into Windows.
- Hiren’s BootCD PE: a WinPE-based rescue USB that includes CrystalDiskInfo, SeaTools for DOS, Victoria, and HDDScan. I keep it on a Ventoy multi-ISO USB alongside my Windows installers.
If a broader hardware issue is suspected (not just the drive), combine drive testing with a full AIDA64 System Stability Test and MemTest86 RAM test.
What to Do When a Drive Fails
- Stop using the drive for anything new. Don’t run defragmentation, don’t run chkdsk with /f, don’t reformat. Every write increases the chance of losing data.
- Clone the drive before copying off files. A sector-level clone (DiskGenius, Clonezilla, or dd) is safer than copying individual files — if a read error happens during the clone, you can rerun just the failed region, and the clone can be worked with offline.
- Copy irreplaceable data off the clone (not the original) to a known-good destination drive.
- Replace the drive. For SSD upgrades, see my guide on the best SSDs for upgrading or replacing your drive.
- If this is your OS drive, migrate Windows — my guide on how to clone or migrate Windows to another drive walks through the full process with DiskGenius.
Frequently Asked Questions
What is Seagate SeaTools and does it work on non-Seagate drives?
SeaTools is Seagate’s free drive diagnostic tool, but it works on any HDD or SSD from any manufacturer — Western Digital, Samsung, Crucial, SanDisk, Kingston, Toshiba, and every other brand. Don’t let the Seagate branding put you off.
Does CrystalDiskInfo work with SSDs?
Yes — CrystalDiskInfo reads SMART data from HDDs, SATA SSDs, and NVMe drives equally well. For NVMe drives it reports the NVMe-specific health attributes (percentage used, data written, media errors) alongside the standard SMART counters.
What does a Caution status in CrystalDiskInfo mean?
It means one or more SMART attributes have crossed their manufacturer-defined warning thresholds — typically reallocated sectors, pending sectors, or uncorrectable errors. Caution is not “might fail later”; it’s “is failing now.” Back up your data immediately and order a replacement drive.
Should I use the Quick Test or the Extended Test?
Quick Test first — it takes 2 minutes and catches obvious failures. If Quick Test passes but you still suspect the drive (intermittent symptoms, SMART warnings), run Extended Test as well. Extended reads every sector on the drive and takes several hours on a mechanical HDD, but it catches bad sectors that a Quick Test would miss.
Can SeaTools actually fix a failing drive?
No — the Fix All button remaps bad sectors to the drive’s spare-sector pool. That buys you time to back up data, but it does not repair the underlying physical fault. A drive that has started reporting errors should be replaced, not repaired. Modern drives are too cheap to risk data loss over.
My drive tests fine but the system is still slow — what now?
If both SeaTools and CrystalDiskInfo come back clean, the fault is elsewhere. Run a full AIDA64 System Stability Test to check CPU, memory, and GPU, or a MemTest86 RAM test to rule out memory. Software causes (Windows bloat, indexing, background services) are also common — Winhance can fix most of those in a few clicks.
