How I Fixed NTFS Mounting Errors on Kubuntu Without Reinstalling Linux

After switching to Kubuntu, I started seeing mounting errors every time I turned on my PC. Two NTFS partitions, named Work and Temporary, refused to open. The error suggested a wrong filesystem type, a bad superblock, a missing helper program, or another mounting problem.

The confusing part? I had used Pop!_OS for around two years without noticing this issue.

Rather than reinstalling Kubuntu or formatting the partitions, I investigated the filesystem errors, repaired the reported NTFS inconsistencies with ntfsfix, and tested normal mounting again. After restarting, both partitions worked and the startup warnings were gone.

This post documents the process, the commands I used, and what I learned along the way.

The original error

Kubuntu displayed errors similar to these:

An error occurred while accessing 'Temporary':
Error mounting /dev/sda6 at /run/media/maruf/9C6A841F6A83F3F4:
wrong fs type, bad option, bad superblock, or other error

The Work partition showed a similar message for /dev/sda3.

My storage layout was:

Kubuntu itself was installed on a separate drive, so the problem appeared to be isolated to two NTFS partitions on the 1 TB disk.

Important: Device names such as /dev/sda3 can change between systems or boots. Always verify the partition layout on your own machine before running repair commands.

Step 1: Identify the affected partitions

I started by listing the filesystems, labels, UUIDs, and mount points:

lsblk -o NAME,SIZE,FSTYPE,LABEL,UUID,MOUNTPOINTS

This confirmed that both affected partitions used NTFS. I then inspected the kernel log for NTFS and disk-related messages:

sudo journalctl -k -b --no-pager \
  | grep -Ei 'sda3|sda6|ntfs|I/O error' \
  | tail -80

The log contained messages like:

ntfs3(sda3): It is recommended to use chkdsk.
ntfs3(sda3): volume is dirty and "force" flag is not set!
ntfs3(sda6): It is recommended to use chkdsk.
ntfs3(sda6): volume is dirty and "force" flag is not set!

These messages pointed to an NTFS filesystem-state problem, rather than just a missing desktop shortcut.

Step 2: Understand the MFT mirror mismatch

I checked both partitions with ntfsfix in no-action mode:

sudo ntfsfix --no-action /dev/sda3
sudo ntfsfix --no-action /dev/sda6

The output included:

$MFTMirr does not match $MFT (record 3).

NTFS uses a Master File Table (MFT) to store metadata about files and directories. It also keeps a mirror of certain critical MFT records. The message indicated that the main MFT and its mirror disagreed about record 3.

The initial checks also reported failed mounting and an attempted correction. That was a useful clue, but it did not mean the filesystems were fully healthy.

A key distinction: ntfsfix is a limited Linux repair utility. It is not a full replacement for Windows chkdsk and cannot repair every possible NTFS inconsistency.

Step 3: Run the basic NTFS repair

After reviewing the results, I ran the repair command on each affected partition:

sudo ntfsfix /dev/sda3
sudo ntfsfix /dev/sda6

Both commands reported that they corrected the difference in $MFTMirr record 3 and completed processing the partitions successfully.

The output also mentioned processing the MFT and its mirror, setting required flags, emptying the journal, and checking the alternate boot sector.

However, the kernel had previously reported that both volumes were dirty. The next step was to address that state and then verify normal mounting.

Step 4: Clear the NTFS dirty flags

I ran:

sudo ntfsfix --clear-dirty /dev/sda3
sudo ntfsfix --clear-dirty /dev/sda6

Both commands completed successfully, reporting:

Mounting volume... OK
Processing of $MFT and $MFTMirr completed successfully.
Checking the alternate boot sector... OK
NTFS partition was processed successfully.

Clearing the dirty flag is not the same as proving that every filesystem structure is healthy. It simply addresses that particular volume state when the utility can do so. This is why I continued with real mounting tests instead of assuming the issue was solved.

Step 5: Verify the filesystem checks

I reran the no-action checks:

sudo ntfsfix --no-action /dev/sda3
sudo ntfsfix --no-action /dev/sda6

This time, both reported successful processing, and neither output showed the earlier MFT mirror mismatch.

I also checked recent kernel messages:

sudo journalctl -k --since "5 minutes ago" --no-pager \
  | grep -Ei 'ntfs|sda3|sda6|I/O error'

At that point, the output still showed older messages from before the repair. That was important: old log entries should not be mistaken for evidence that the latest repair failed.

Step 6: Test normal mounting

Instead of forcing either partition to mount, I used udisksctl, which asks the system's disk-management service to mount the volume normally:

udisksctl mount -b /dev/sda3
udisksctl mount -b /dev/sda6

Both succeeded:

Mounted /dev/sda3 at /run/media/maruf/5D3C6A6A1D3BE7F7
Mounted /dev/sda6 at /run/media/maruf/9C6A841F6A83F3F4

That confirmed that Kubuntu could mount both partitions normally. I then restarted the PC to test the original startup problem.

The result

After restarting Kubuntu, both Work and Temporary worked normally, and the startup mounting warnings were gone.

No reinstall was needed. I did not format the partitions, change the partition table, force-mount the volumes, or modify /etc/fstab.

The repair commands addressed the reported MFT mirror mismatch and dirty-volume state, and normal mounting worked afterward. We cannot determine from this troubleshooting session whether switching from Pop!_OS caused the original inconsistency or simply exposed it.

Lessons learned

• Read the logs before changing things. lsblk and journalctl helped narrow the issue down to the affected NTFS partitions.
• Understand the error before repairing. The MFT mirror message gave a more specific clue than the generic desktop mounting dialog.
• Use ntfsfix with realistic expectations. It can resolve some common NTFS issues, but it cannot perform every repair that chkdsk can.
• Verify with a normal mount. A successful command is encouraging, but actually mounting the partitions and checking the files is essential.
• Test after restarting. The original issue happened at startup, so a reboot was the final practical test.
• Back up important files. A successful mount does not guarantee that every part of a filesystem is free of corruption.

What if the problem comes back?

If the MFT mismatch or dirty-volume warning returns, avoid repeatedly clearing the dirty flag or using force-mount options. Back up accessible data first and investigate the filesystem and disk health.

Because I did not have Windows installed, a full Windows chkdsk repair was not immediately available. If Linux-side repair is insufficient, a Windows recovery environment or another carefully planned recovery approach may be necessary.

Final thoughts

What initially looked like a reason to reinstall Kubuntu turned out to be a repairable NTFS mounting problem. By inspecting the partition layout, reading the kernel logs, using ntfsfix, and testing normal mounting after a restart, I restored access to both partitions without formatting or reinstalling the operating system.

The biggest lesson: diagnose first, repair carefully, and verify the result before making permanent configuration changes.