I am running Unraid 7.3.0 with 1 12TB Parity Disk and 8 12TB Data Disks , 1 NVME for Docker Apps and a USB Boot drive. I have been running it on trial with 2 extensions already and my trial expires in 5 days. I was planning on purchasing it anyways since I have been very happy with how it works and what I have learned thus far. Also setting up docker apps and shares was much easier than TrueNas (which I tried for a bit).

A few days ago, I was re-checking a file with had error-ed in qBit when my shares stopped responding. I stopped the array and attempted to restart the array, but the webui refreshed but the array wouldn't start. I restarted the PC (which I now know not to for diagnostic purposes). Everything restarted normally. Once back up I attempted to start the array but got "unmountable, no filesystem" for disk 1. I ran xfs_repair (as previous posts here suggested) and it found errors. I then hit the "fix" button to do an actual repair. On the second xfs_repair attempt i ended up with the error.

entry "The Comeback - S03E02 - Valerie Has a Secret.mkv" at block 0 offset 512 in directory inode 1457253450 references free inode 757426606
	clearing inode number in entry at offset 512...
bad data fork in inode 21559458520
rebuilding inode 21559458520 data fork

fatal error -- adding rmap to in-memory btree, err 117

I have been unable to find much on this error, just that it may be tied to not enough ram. To be safe, I went ahead and ran memtest from the unraid boot menu on my ram and it passed multiple passes. I then ran the xfs_repair again, but i still get the same error and stops at the same place as before.

I did tried to run xfs_repair with the -m option to increase ram via console. I even went as high as 10240M, so 10gb. And still get the same error. I do notice that the amount of RAM being used according to the dashboard never goes up when running the repair regardless of the -m setting. I have 16GB of DDR4 total.

I ordered a new disk which will be arriving next week. I was thinking of replacing it and letting the array rebuild it, then trying again to repair if needed but someone else posted here recently with this same exact error and had no success with a new disk.

The emulated contents of disk 1 also seem to show no files when disk 1 is disabled. I have not touched parity or any other drives. So I am afraid parity would not help me here.

Anyone have any clues on this error? There is very little info about it online from what i can find. I have attached my diagnostic file.

jesus-server-diagnostics-20260515-0307.zip

If xfs_reapir cannot fix the fileystem and you don't have backups, you can try a file recovery app, like UFS explorer, then reformat the disk.

That's what I am afraid of. Luckily the data was spread across multiple disks due to how unraid by default only fills half the drives before moving on to the next. I am going to wait till I get the new disk. Put that in and attempt to recover any important files from the old disk. When I put the new disk in will unraid attempt to rebuild the array? Should I let it?

6 hours ago, Thamian Solus said:

When I put the new disk in will unraid attempt to rebuild the array? Should I let it?

You can, or you will need to resync parity. Either way, the new disk will need to be formatted in the end.