Yeah, this one's annoying. Let's get you back to work.
You're seeing ERROR_EA_LIST_INCONSISTENT (0x000000FF) — Windows is telling you the extended attributes (EAs) on a file or folder are broken. The culprit here is almost always a failed backup, a botched robocopy with the /COPYALL flag, or someone yanking a USB drive mid-write. I've fixed this on everything from Windows 10 workstations to Server 2019 file shares. Here's the playbook.
The Direct Fix: chkdsk + fsutil
Don't bother reinstalling or running SFC first — it rarely helps with EA corruption. The real fix is a two-step process. Run this as Administrator.
chkdsk C: /f
This scans the volume for file system errors and repairs them. If the drive is the system volume, it'll schedule a scan on next reboot. Reboot and let it run.
After chkdsk finishes, run this to clear the broken EA list:
fsutil repair query C:
If that reports Repair level: 0, you're good. If it shows a non-zero value, force a repair with:
fsutil repair set C: 1
chkdsk C: /f
Repeat until the repair level hits zero. Then try accessing the file or folder that triggered the error again.
Why This Works
Extended attributes are a hidden part of NTFS — they store metadata like file classification tags, encryption info, or app-specific data. When they get inconsistent, the file system doesn't know which version of the EA list to trust. chkdsk /f rebuilds the NTFS metadata, and fsutil repair forces the volume health checker to re-evaluate and fix the EA structure. It's like resetting a garbled cache.
This works because the EA data itself is usually intact — the index pointing to it got scrambled. The repair tools just reindex the attributes.
Less Common Variations of This Error
EA Error on a Network Share
If you see 0x000000FF trying to copy a file from a NAS or another Windows server, the EAs are corrupt on the source volume. You can't fix a remote volume from your machine. You need to run the chkdsk + fsutil combo directly on the server hosting the share. Remote desktop in, or use psexec to run the commands.
EA Error During robocopy with /COPYALL
robocopy /COPYALL copies all attributes, including EAs. If the source has corrupt EAs, robocopy will fail with this error. Workaround: use /COPY:DAT instead — it skips EAs. Then fix the source volume later. Example:
robocopy C:\source \\server\share /E /COPY:DAT
EA Error After a Backup Restore
Some backup tools (looking at you, older versions of Veeam and Windows Server Backup) restore EAs incorrectly. If the error appears right after a restore, delete the restored file and re-restore it without EA support. Check your backup software's settings for an "ignore extended attributes" flag.
EA Error on a USB Drive
Removable drives formatted as NTFS can get EA corruption if ejected improperly. Run chkdsk on the drive letter directly:
chkdsk E: /f
Then try to delete or access the file. If that fails, reformat the drive (backup data first).
Prevention
Three things will stop 95% of these errors:
- Always safely eject USB drives. I know, it's a pain. But pulling an NTFS drive without clicking "Eject" is asking for EA trouble.
- Use
/COPY:DATinstead of/COPYALLin robocopy unless you absolutely need EAs. Most people don't. You'll know if you do. - Schedule periodic chkdsk scans on file servers. Run
fsutil dirty query D:weekly to catch issues early. A quickchkdsk /fonce a month on non-system volumes prevents corruption from compounding.
One last thing: if you're on a RAID array and seeing EA errors regularly, check your storage controller drivers and firmware. I've seen flaky RAID controllers corrupt NTFS metadata under load. Update those first before blaming Windows.