How to Check Hard Drive Health: SMART Tests, Bad Sectors & Warning Signs
Quick Answer
Open CrystalDiskInfo (free, Windows) and check the health status shown in the large colored indicator. Blue means Good; Yellow means Caution; Red means the drive has failed thresholds and needs immediate replacement. The three SMART attributes that matter most for hard drives are ID 05 (Reallocated Sectors), ID C5 (Pending Sectors), and ID C6 (Uncorrectable Sector Count) — all three should read zero. Any non-zero value signals active failure.
Hard drives are the one component most likely to fail catastrophically and take data with them. Unlike SSDs, HDDs give you audible and measurable warnings before failure — if you know what to listen for and how to read SMART data. A few minutes of diagnostic work every three to six months can be the difference between a planned migration and an emergency data recovery bill.
How Hard Drive SMART Works
SMART (Self-Monitoring, Analysis and Reporting Technology) is built into every modern hard drive. The drive's firmware continuously logs internal metrics — read error rates, spin-up time, reallocated sector counts, temperature, shock events — and stores them in a small onboard register. Diagnostic software reads this register and compares values against manufacturer-set thresholds to determine if the drive is within normal operating parameters.
Hard drives benefit from SMART monitoring more than SSDs because their failure modes are more predictable. A hard drive with rising bad sector counts is almost always on a trajectory toward failure — the question is timing, not whether.
The Three SMART Attributes That Predict HDD Failure
| ID | Attribute Name | What it means |
|---|---|---|
| 05 | Reallocated Sectors Count | Sectors the drive has moved to spare area because they were damaged. Any value above zero means the drive has encountered read failures. |
| C5 | Current Pending Sector Count | Sectors flagged for reallocation on the next successful write. These are sectors the drive cannot currently read reliably. |
| C6 | Offline Uncorrectable Sector Count | Sectors that could not be read during an offline scan and for which error correction also failed. These represent permanent data loss on those sectors. |
All three should be at zero on a healthy drive. A single non-zero value in any of these three is a serious warning — back up immediately and begin planning replacement, even if the drive otherwise seems to work normally.
How to Check Hard Drive Health on Windows
CrystalDiskInfo (Free, Recommended)
- Download CrystalDiskInfo from crystalmark.info and run it.
- Each drive appears as a tab. The large colored status indicator shows: Blue = Good, Yellow = Caution, Red = Bad/Failed.
- Scroll the attribute list to find IDs 05, C5, and C6. Verify their "Current" values and "Raw Value" are both 0.
- Check "Power On Hours" for drive age context, and "Temperature" — HDDs should operate below 55°C under load.
HD Tune (Free Version — Adds Error Scan)
HD Tune's Error Scan maps every sector on the drive. Green blocks are readable; red blocks are unreadable. Any red blocks indicate bad sectors and match directly to SMART attributes C5 and C6. Run the scan in the background — it takes 20–60 minutes depending on drive size.
Manufacturer Diagnostic Tools
- Seagate: SeaTools for Windows — runs short and long self-tests
- Western Digital: WD Dashboard — Data Lifeguard diagnostics
- Toshiba: Toshiba Storage Diagnostic Tool
Manufacturer short tests run in 1–2 minutes. The extended test (often called "Long Test" or "Full Test") scans every sector and takes several hours but is the most thorough check available.
How to Check Hard Drive Health on macOS
- Hold Option and click the Apple menu → System Information.
- Click Storage in the left sidebar.
- Select the hard drive and look for SMART Status: Verified. If it says "Failing," back up and replace immediately.
For full SMART attribute access on macOS, install smartmontools via Homebrew:
brew install smartmontools
sudo smartctl -a /dev/disk0How to Check Hard Drive Health on Linux
sudo smartctl -a /dev/sda # Full SMART data
sudo smartctl -H /dev/sda # Pass/fail summary only
sudo smartctl -t short /dev/sda # Run a short self-test
sudo smartctl -t long /dev/sda # Run an extended self-test (hours)After a self-test, read the results with sudo smartctl -l selftest /dev/sda. Any "Completed: read failure" or "Failed!" results mean the drive has sectors it cannot read.
What Bad Sectors Mean — and What to Do
A bad sector is a physical location on the disk platters that the drive can no longer reliably read or write. Hard drives have a pool of spare sectors reserved for exactly this purpose — when a bad sector is detected, the drive remaps data to a spare and flags the original location. This is Reallocated Sectors Count going up.
A few reallocated sectors on an older drive are not immediately fatal, but the count should never increase. If you recheck in two weeks and the number has grown — even by one — the drive is actively failing and should be replaced before the spare sector pool runs out.
Sound-Based Warning Signs
Hard drives make distinctive sounds before failure. Learn to recognize them:
- Clicking or ticking (click of death): The read/write head is failing to find its calibration track. This is a severe failure precursor — back up immediately.
- Grinding or scraping: Head-to-platter contact. The drive is in the process of failing. Stop using it and recover data.
- Repeated spin-up attempts: Motor or bearing issue. The drive may still work intermittently, but failure is imminent.
- Unusual buzzing: May indicate a bearing problem, especially if accompanied by vibration.
Any unusual mechanical sound from a hard drive is a reason to back up before running any diagnostic — the act of running tests can sometimes accelerate failure on a drive that is already physically compromised.
FAQs
Is it safe to keep using a hard drive with one bad sector?
One reallocated sector on an older drive is not catastrophic, but you should recheck in 2–4 weeks. If the count has increased, the drive is actively degrading — back up and replace. A rising count is more dangerous than a static count, regardless of the absolute number.
How long do hard drives typically last?
Consumer HDDs typically last 3–5 years in continuous use. Enterprise drives are rated for 5–7 years. However, individual drives vary widely — some fail in year one, others run for 10+ years. Use SMART monitoring and regular backups rather than relying on age as a predictor.
Does a SMART PASSED result mean my hard drive is healthy?
Not necessarily. SMART PASSED means no individual attribute has crossed its manufacturer threshold. But the thresholds are often set conservatively — a drive can have several reallocated sectors and still pass the overall test. Always review individual attribute values, especially IDs 05, C5, and C6.
How often should I check hard drive health?
Every 3–6 months for a healthy drive, monthly if any SMART attribute has shown a Caution reading. Set CrystalDiskInfo to run at startup in resident mode — it sits in the system tray and alerts you if any attribute crosses a threshold.
Can I recover data from a failing hard drive?
Often yes, if the drive can still spin up and the platters are not physically damaged. Use data recovery software (Recuva, R-Studio) for logical errors. For drives with clicking sounds or physical damage, professional data recovery services can often recover data but cost $300–$1,500+. Act quickly — continuing to use a failing drive reduces recovery chances.
Summary
Hard drive health checks take two minutes with CrystalDiskInfo or smartctl. The three attributes that matter most — Reallocated Sectors (ID 05), Pending Sectors (ID C5), and Uncorrectable Sectors (ID C6) — should all be zero on a healthy drive. Any non-zero reading warrants immediate backup and replacement planning. Unusual mechanical sounds, especially clicking or grinding, are emergency warnings that demand action before any diagnostic.