How to Run chkdsk to Fix Disk Errors in Windows 11

What chkdsk fixes — and what it does not

chkdsk operates on the file system — NTFS, FAT32 or exFAT — checking the master file table, the security descriptors and the directory tree for inconsistencies. It fixes what /F (or the Windows GUI equivalent) finds; /R extends the same pass with a sector-by-sector read of the whole surface, flagging unreadable clusters and moving their data to spare ones where possible. A failing disk, however, is not a file system problem: if SMART reports falling attributes or bad sectors keep multiplying, chkdsk is first aid, not the cure.

Windows also maintains the “dirty bit” — an internal flag set when the file system was left inconsistent, typically after a power loss or a hard freeze. At the next boot Windows notices the flag and runs an automatic chkdsk; fsutil dirty query C: shows you whether the flag is currently set on a volume.

• chkdsk C: /f — fixes file system errors; on the system drive it is scheduled for the next reboot
• chkdsk C: /R — everything /f does plus the bad-sector surface scan; hours on a large HDD
• chkdsk D: /f on a data drive — prompts to dismount the volume first if files are open
• chkdsk C: /scan — an online read-only check that reports problems without fixing them
• fsutil dirty query C: — shows whether the volume is flagged for a boot-time check

Running chkdsk: GUI and command line

The GUI route covers everyday cases: open File Explorer → This PC, right-click the drive → Properties → Tools → Error checking → Check. Windows 11 may answer that the drive is fine; choose Scan drive anyway to force a pass. This check runs online, reports issues and, when repairs are needed, offers to schedule them for the next restart.

The command line is where the real repair happens: open Terminal (Admin), run chkdsk C: /f, answer Y to schedule it, and reboot. For a data drive use its letter instead and allow the dismount when prompted. Reach for /R only with evidence of bad sectors — random read errors, files that vanish, SMART warnings — because an SSD's controller remaps weak cells internally and the hours-long scan rarely adds information. And never interrupt a running pass: a power cut mid-chkdsk is a reliable recipe for exactly the damage it was meant to fix.

Reading the results and what to do next

After a boot-time check the results are not shown in a window — they land in Event Viewer: Windows Logs → Application, source Wininit, event 1001. The entry lists what was checked and what was fixed, and its absence after a scheduled run usually means the dirty bit cleared without repairs. A clean pass on a healthy HDD is normal; recurring bad sectors on the same disk month after month is the disk asking to be replaced.

Two follow-ups make sense after any serious repair: check the disk's SMART health so you know whether the errors were logical or physical, and clear the gigabytes of junk a troubled machine accumulates — failed update files, temp folders, crash dumps — so the next backup is smaller and faster. Kleaner PRO automates that cleanup on a schedule. If the SMART check is already failing, stop cleaning and start copying the data off first.

Questions and Answers

Read next

All articles · FAQ