How to Check Hard Drive Health in Windows Server

To check hard drive health in Windows Server, start with an elevated PowerShell session and review physical-disk status. Then inspect reliability counters and, when the hardware exposes them, SMART data. Use chkdsk separately for file-system errors; it is not a hardware health test.

Disk Management helps identify disks, partitions, and volumes. A volume marked “Healthy” does not confirm that its physical drive passed SMART diagnostics. See Microsoft’s Disk Management overview.

Before you check a server disk

If you hear new clicking or grinding, see repeated I/O errors, or notice a sudden slowdown, verify that a recent backup is usable before repairs or extended tests. Record the drive model, serial number, bay, and dependent volumes so you can identify the right device.

RAID controllers, HBAs, virtual machines, and some drivers can hide physical disks or SMART data. A virtual machine may see only a virtual disk. Check the server or controller management utility, or the hypervisor host, alongside Windows. Microsoft’s Windows Server storage troubleshooting guidance recommends checking storage drivers and firmware with the hardware vendor.

1. Review physical-disk status in PowerShell

Open PowerShell as an administrator and run this read-only command:

Get-PhysicalDisk | Format-Table FriendlyName, SerialNumber, MediaType, Size, OperationalStatus, HealthStatus -AutoSize

Get-PhysicalDisk lists disk objects visible to available Storage Management Providers. Match the model or serial number to the drive you intend to inspect. HealthStatus can be Healthy, Warning, Unhealthy, or Unknown; OperationalStatus reports its current operating state. These provider-reported signals cannot guarantee a drive will not fail. Investigate warnings, missing disks, or unexpected states, especially in a RAID set or Storage Spaces pool.

2. Read the reliability counters

Replace PhysicalDisk8 below with the friendly name shown on your server:

Get-PhysicalDisk -FriendlyName "PhysicalDisk8" | Get-StorageReliabilityCounter | Format-List

Get-StorageReliabilityCounter can return temperature, power-on hours, wear, and corrected or uncorrected read/write errors. Some fields may be blank because the device, driver, or controller does not expose them; blank does not mean there is no problem.

Compare trends with the drive or server manufacturer’s guidance. There is no universal temperature, wear, or error-count cutoff for every model. If warnings appear, errors rise, or an expected drive is missing, protect the data and inspect the vendor or controller diagnostics.

3. Cross-check SMART and controller diagnostics

For a graphical SMART view, download GSmartControl from its official downloads page. It is a graphical interface for smartctl that can display SMART attributes and offer self-tests for supported drives. The project lists SATA, PATA, and NVMe devices, plus some drives behind USB bridges or RAID controllers. Confirm compatibility with your Windows Server edition and software policy before installing it on a production server.

Select the intended drive and review its identity, overall assessment, and attributes. A passing summary cannot guarantee that a drive will not fail. If a controller hides SMART data, use the server or controller vendor’s approved diagnostics. Run a self-test only when appropriate for your hardware and maintenance window.

4. Check the NTFS file system with CHKDSK

To scan a local NTFS volume online, open an administrator Command Prompt and replace D: with the correct volume letter:

chkdsk D: /scan

Microsoft documents /scan as an online NTFS scan. It checks file-system structures and metadata, not SMART health or the drive’s mechanical or electronic condition. If it finds corruption, back up the volume and plan a repair. Options such as /f and /r make changes and can require the volume to be locked or checked during restart; review the CHKDSK command reference before scheduling repairs on a production server.

How to act on the results

• Healthy or OK: The provider reports no current warning. Keep routine backups and monitoring; this is not a guarantee against failure.
• Warning, Unhealthy, Unknown, increasing errors, or a missing disk: Back up critical data and inspect vendor/controller diagnostics and recent system events. Check array or Storage Spaces health before replacement; do not pull or reset a pooled disk without following its replacement procedure.
• CHKDSK reports a file-system problem: Plan an approved volume repair, and check hardware health separately. File-system errors alone do not identify their cause.

For a production server, compare Windows’ report with the vendor or controller view. Confirm backups and redundancy before replacing a drive, then monitor the system after repair or replacement.