Fix NTFS error 0XC0000278: reparse data invalid

This error means Windows found corrupt reparse point data on an NTFS drive. The fix is to delete the bad reparse point via fsutil, then chkdsk the volume.

You're staring at an error that says the user data passed for a reparse point is invalid — 0XC0000278. Yeah, that's annoying. Windows is refusing to access a file or folder, and it's not telling you which one. Let's fix it.

The fix: delete the bad reparse point

The only reliable way to resolve this error is to identify the corrupt reparse point and delete it. Here's the step-by-step:

Open Command Prompt as Administrator. Don't skip this — without admin rights, fsutil won't let you touch reparse points.
Run chkdsk X: /scan where X is the drive letter showing the error. This will scan the volume without taking it offline. If you can't access the drive at all, use chkdsk X: /f instead — but that requires a dismount or a reboot.
Check the chkdsk output for lines like Reparse point in file ### is corrupt. Write down the file index number (the decimal number after "file").
Find the actual file path using fsutil reparsepoint query X: /index ###. Replace ### with the index number from step 3.
Once you have the full path, run fsutil reparsepoint delete "X:\path\to\file". If it's a folder, include the trailing backslash.
Run chkdsk X: /scan again to confirm the error is gone.

That's it. The error disappears because you removed the invalid reparse data that NTFS couldn't resolve.

Why this works

Reparse points are special NTFS metadata tags that tell Windows to redirect file operations to a different handler — think symbolic links, junction points, or OneDrive placeholder files. When the data inside a reparse point gets corrupted (a partial write, a bad sector hitting that metadata, or a botched tool like a migration utility), NTFS can't interpret it. It returns STATUS_IO_REPARSE_DATA_INVALID — error 0XC0000278.

The fsutil command doesn't try to fix the corrupt data. It removes the entire reparse point, leaving the underlying file or folder intact. After that, the file is just a normal file. Windows can access it again. Chkdsk then verifies the file system consistency around that spot.

Less common variations

Sometimes the error pops up during a robocopy or backup operation but not during normal use. That's because the backup tool reads reparse point data that Explorer skips. In that case, the same fsutil delete command fixes it.

Another variation: the error appears on a mounted VHD or VHDX. The fix is identical — run chkdsk inside the virtual disk, find the corrupt reparse point, delete it. But you might need to diskpart and attach vdisk first if the volume doesn't show up normally.

If chkdsk reports no corrupt reparse points but the error persists, the corruption might be in the $MFT itself. Run chkdsk X: /r to scan for bad sectors and recover readable data. This is rare — 0XC0000278 is almost always tied to a specific reparse point.

Prevention

Most of the time, this error comes from unexpected power loss while NTFS is writing reparse point metadata. A UPS on your machine stops that cold.

If you use tools like mklink for junctions or OneDrive with Files On-Demand, be careful about moving the target directory while the link is active. Move the link first, then the target. Also, avoid running chkdsk on a drive while another process is writing to it — that's a classic way to corrupt reparse data.

Regular file system health checks with fsutil dirty query X: can catch a dirty bit early. Clean it with chkdsk X: /f at boot, before any corruption spreads.