How to check SSD and hard drive health on Windows, macOS and Linux
Drives rarely fail without warning. SSDs and hard drives record their own health in S.M.A.R.T. data: error counts, temperature, power-on time and, for SSDs, how much of their write endurance is used. A few minutes with the right tool tells you whether a drive is fine or whether it is time to back up and replace it.
Windows: CrystalDiskInfo
- Download and start CrystalDiskInfo.
- Each drive appears as a tab at the top. The large box shows the Health Status: Good, Caution or Bad, and for many SSDs a percentage of remaining life.
- Check the temperature and the attribute table below. For hard drives, watch reallocated, pending and uncorrectable sector counts; for SSDs, the wear or "percentage used" values and media errors.
Caution does not always mean imminent failure, but it does mean you should back up important data now and keep an eye on the drive.
Windows: HWiNFO and built-in tools
HWiNFO shows drive temperatures and S.M.A.R.T. values alongside the rest of your hardware, which is handy when you are also watching CPU and system temperatures. Windows also reports a basic health state in PowerShell:
PowerShell (run as administrator)
Get-PhysicalDisk | Select-Object FriendlyName, MediaType, HealthStatus
Get-PhysicalDisk | Get-StorageReliabilityCounter | Select-Object DeviceId, Temperature, Wear, ReadErrorsUncorrectedOn Windows 11, Settings > System > Storage > Advanced storage settings > Disks & volumes > Properties can show drive health information for NVMe SSDs.
macOS
Open Disk Utility, select the physical drive and look at the S.M.A.R.T. status, which reads Verified when the drive reports no problems. For detailed attributes, use a S.M.A.R.T. tool such as smartmontools.
Linux
The smartmontools package provides smartctl. Replace /dev/sda with your drive (use lsblk to list them); NVMe drives are named like /dev/nvme0n1:
Terminal
sudo smartctl -H /dev/sda
sudo smartctl -a /dev/sdaThe first command prints an overall PASSED or FAILED assessment; the second shows every attribute and the drive’s error log.
If the drive looks unhealthy
Back up first, before running long tests or repairs, because heavy activity can push a failing drive over the edge. To copy a whole system drive to a new one, Clonezilla can image or clone it, and GParted can adjust partitions afterwards. More tools are listed in our disk and USB utilities guide.
Tools used in this guide
-
Reads S.M.A.R.T. data and shows a clear Good, Caution or Bad rating with temperature for every drive.
-
Detailed hardware information and live sensors, including drive temperatures and health values.
-
Image or clone a drive to a new one before an ageing disk fails.
-
Resize and move partitions after cloning to a larger drive.
Frequently asked questions
How often should I check drive health?
A quick look every few months is enough for most people, plus whenever a PC becomes unusually slow, freezes or makes new noises.
Is a high power-on hours value a problem?
Not by itself. Error counts, reallocated or pending sectors and SSD wear are better indicators than age alone.
Does checking S.M.A.R.T. data wear out an SSD?
No. Reading health data does not write to the drive.
CrystalDiskInfo says Caution. Should I replace the drive?
Back up immediately and look at which attribute triggered the warning. A few reallocated sectors may stay stable for a long time, but growing numbers are a strong sign to replace the drive.
More guides
How to create a bootable USB drive (with Ventoy) · How to verify a SHA-256 checksum (Windows, macOS, Linux)