Alright, I have a somewhat lengthy story... TL;DR: any ideas for why my array doesn't want to mount an NTFS drive in my array but it will mount the partition and show files in the terminal?

So I had a 10tb NTFS drive that came with files on it I wanted to keep (which is why it was NTFS), a 1tb xfs drive with some added files, and a 22 tb parity drive in an array with valid parity.

I was swapping/upgrading towards a 24tb parity + 22tb drive for additional space and to get the 1tb and probably 10tb out of there. Additionally, I wanted to move the data drive from ntfs to xfs. In retrospect, given the change in filesystem, I think I should have just created a new config, copied data from manually mounted previous drives to the "new" 22tb data xfs array and let the parity build as data made it onto the new array.

Instead what I did was:

1. replace the 22tb parity with the new 24tb drive

2. rebuild parity based on 10tb and 1tb

3. swap out the 10tb for the 22tb

4. (at this point I realized that I was going to rebuild the NTFS fs, and tried to reload the 10tb drive into the system in maintenance mode, but it didn't work).

5. do a data rebuild on the 22tb drive in maintenance mode

6. restart the array in normal mode

And I get the message that the disk is unmountable, no or wrong file system.

Here are some logs:

Jan 11 19:43:19 RaidStar root: Checking the alternate boot sector... OK

Jan 11 19:43:19 RaidStar root: NTFS volume version is 3.1.

Jan 11 19:43:19 RaidStar root: NTFS partition /dev/md1p1 was processed successfully.

Jan 11 19:43:20 RaidStar emhttpd: shcmd (2340): ntfsresize -f -P /dev/md1p1

Jan 11 19:43:20 RaidStar root: ntfsresize v2022.10.3 (libntfs-3g)

Jan 11 19:43:23 RaidStar root: Device name : /dev/md1p1

Jan 11 19:43:23 RaidStar root: NTFS volume version: 3.1

Jan 11 19:43:23 RaidStar root: Cluster size : 4096 bytes

Jan 11 19:43:23 RaidStar root: Current volume size: 22000969916928 bytes (22000970 MB)

Jan 11 19:43:23 RaidStar root: Current device size: 22000969920512 bytes (22000970 MB)

Jan 11 19:43:23 RaidStar root: New volume size : 22000969916928 bytes (22000970 MB)

Jan 11 19:43:23 RaidStar root: Nothing to do: NTFS volume size is already OK.

Jan 11 19:43:23 RaidStar emhttpd: shcmd (2341): ntfsfix -d /dev/md1p1

Jan 11 19:43:23 RaidStar root: Mounting volume... OK

Jan 11 19:43:23 RaidStar root: Processing of $MFT and $MFTMirr completed successfully.

Jan 11 19:43:23 RaidStar root: Checking the alternate boot sector... OK

Jan 11 19:43:23 RaidStar root: NTFS volume version is 3.1.

Jan 11 19:43:23 RaidStar root: NTFS partition /dev/md1p1 was processed successfully.

Jan 11 19:43:23 RaidStar emhttpd: shcmd (2342): mount -t ntfs3 -o noatime,acl /dev/md1p1 /mnt/disk1

Jan 11 19:43:23 RaidStar kernel: ntfs3: Enabled Linux POSIX ACLs support

Jan 11 19:43:23 RaidStar kernel: ntfs3: Read-only LZX/Xpress compression included

Jan 11 19:43:23 RaidStar root: mount: /mnt/disk1: wrong fs type, bad option, bad superblock on /dev/md1p1, missing codepage or helper program, or other error.

Jan 11 19:43:23 RaidStar root: dmesg(1) may have more information after failed mount system call.

Jan 11 19:43:23 RaidStar emhttpd: shcmd (2342): exit status: 32

Jan 11 19:43:23 RaidStar emhttpd: shcmd (2343): rmdir /mnt/disk1

Jan 11 19:43:23 RaidStar kernel: ntfs3(md1p1): NTFS 20489.99 Gb is too big to use 32 bits per cluster.

Jan 11 19:43:23 RaidStar kernel: ntfs3(md1p1): try to read out of volume at offset 0x14027fff2e00

Jan 11 19:43:23 RaidStar emhttpd: disk1: mount error: wrong or no file system


Interestingly, if I open the terminal, I can mount the filesystem by running

mount -t ntfs /dev/sdc "/tmp/sdc1/"

Any ideas? I am also able to access the drive contents if I just plug it in via a usb to sata controller to my linux PC. Any way I can get this drive to mount or commands I can try to fix it up?

I am currently scanning the old 10tb drive because it does not come up with files on my linux or windows pc, but it seems a tool is able to find files through some scanning process. Once that scanning is done and I can save the scan for future recovery, I was also going to try resetting the config and putting the 10tb and 1tb back in and seeing if the array will just find all the files at that point.

Thanks in advance for any suggestions!

Attach Diagnostics ZIP to your NEXT post in this thread.

7 hours ago, jtbrauer said:

Jan 11 19:43:23 RaidStar kernel: ntfs3(md1p1): NTFS 20489.99 Gb is too big to use 32 bits per cluster.

This appears to be the problem, and if it mounts outside the array, it could be related to the md driver using 512B sector sizes vs. 4K outside the array.

My recommendation would be to mount the disk with UD or using the CLI and copy the data, then reformat it with a Linux fs, NTFS has limited support in Linux.

Thanks! I'm recovering files off the 10TB using TestDisk and will check them against what I have on the 22 TB and get everything onto xfs.