I messed around a noticed the other day that I had at least 3 folders in my Pictures archive missing, 2023,2024, and 2025. I had a drive having problems, so I assumed that was the issue, but not so sure now.

I have 15 drives in my array, so the odds of all 3 of those folders being 100% on the failing drive would be nearly impossible as every other folder I have is spread across multiple drives. I even started the array without the drive, and the three folders showed up in the shares, but then after a refresh they disappeared. So I'm scratching my head.

I did a full data recovery on the drive, recovered tons and tons of data, but none that seems to be from those 3 folders. So I am really confused.

I've seen a single folder with little data wind up on a single drive, but the majority of my folders I see get split across multiple drives, so I am confused as to how all three of these folders would wind up together on the same drive, and not a single file from any of the 3 being recovered?

Any thoughts?

You are likely to get more informed feedback if you attach your system's diagnostics zip file to your next post in this thread.

It is always a good idea to do this to allow us to see the current state of your system and so we can see logs.

19 hours ago, mrmagoo_83 said:

I've seen a single folder with little data wind up on a single drive, but the majority of my folders I see get split across multiple drives, so I am confused as to how all three of these folders would wind up together on the same drive, and not a single file from any of the 3 being recovered?

First, your Diagnostics file will help us to solve your problem.

Now, one thing to realize. Windows and Linux handle file (and directory/folder) names differently. Windows ignores 'case' and Linux does not. The file names Test.txt and test.txt are two different files to Linux. Windows would not allow you to create which ever one is created later. The problem arises when the files are created under Linux and then you look at them from Windows using SMB. Windows will only see one of the two and that choice is somewhat obituary from your Windows perspective! If you are looking for this problem, you have to do from the Linux side of the equation. Use the built-in Dynamix File Manager, Krusader, Midnight Commander (mc), or the CML ls -al command.

Edited July 15, 2025Jul 15 by Frank1940

I was looking within the Unraid UI and terminal for these folders. They were there, full of data, then they were not. Very confused by it.

The server is running a rebuild right now, which may blow everything up, not really sure, but when I put the 12TB drive back in it started, I have little faith it will restore anything though.

Once that is done, if it fails to restore the folders, then I'll run diagnostics again.

It just makes zero sense to "lose" 3 folders at once.

diagnostics-20250704-2312.zip

18 hours ago, mrmagoo_83 said:

was looking within the Unraid UI and terminal for these folders. They were there, full of data, then they were not. Very confused by it.

@JorgeB I looked at his diagnostics and I think that the Title of the Thread is not an indicator of what his true problem is. His log file has way too much going on for it to be a simple file management issue. Could you have a look and see what you think.

In the end the array was started in maintenance mode, post new diags in normal mode, but before, disk15 wasn't mounting.

Correct, I was trying to mount a disk I had not formatted yet, that was my fault. But all of that was in efforts to find these 3 missing folders.

I installed the original, correct 12TB drive, the array spent the last two days 'rebuilding' the drive...to nothing. The drive is blank after the rebuild.

Still zero sign of the three folders with my pictures in them.

41 minutes ago, mrmagoo_83 said:

The drive is blank after the rebuild.

Did you format the disk during or before the rebuild?

I am not sure how to answer that question. I thought the drive was cleared as part of the rebuild process before it started reassembling it from parity? I did not force format the drive if that is the question.

2 minutes ago, mrmagoo_83 said:

am not sure how to answer that question. I thought the drive was cleared as part of the rebuild process before it started reassembling it from parity?

No, the drive is written with the content from parity together with the other disks. If the disk is mounted and empty, it was formatted at some point.

But even if that is the case, shouldn't the rebuild have used the data from parity and elsewhere to rebuild the data on the drive? Isn't that how it works if I had replaced that drive with a fresh one?

It does show 250GB used, which is a lot to have no files on it, is it not?

Edited July 16, 2025Jul 16 by mrmagoo_83

14 minutes ago, mrmagoo_83 said:

It does show 250GB used, which is a lot to have no files on it, is it not?

One way to check what is there is to use the built-in Dynamix File Explorer.

Open SHARES (Step 1) and then (step 2) click on the proper icon in the circled area for the disk where you want to look for files.

48 minutes ago, mrmagoo_83 said:

shouldn't the rebuild have used the data from parity and elsewhere to rebuild the data on the drive?

Not if the disk was formatted; formatting a disk deletes all its data and updates parity to reflect that.

2 hours ago, mrmagoo_83 said:

I did not force format the drive if that is the question.

At any point did you check the box at the bottom of the Main page that says "Format - Yes I want to do this" and start the array?

Negative. I have not checked that box.

And yes that is the method I used to go in and view what was 'on' the drive. And all it has is "Parent Directory".

I am confused though, what did it rebuild if it wasn't rebuilding from parity what should have been on the drive? Isn't that how it updates when I replace a drive?

And how is it every other folder I have shows up across multiple drives, but these three latest folders were all on this single drive? That seems very odd to me, shouldn't there be some lingering of the contents on a different drive?

Which disk are we talking about? The diags don't show the rebuild.

I'll get back to y'all. I've been traveling for work.

I have currently shut the server down and I am pulling each drive and doing a recovery attempt on each one to pull off both the current active files, but as well as anything lingering that hasn't been fully wiped.

Using a Linux recovery program and not actually doing anything to the device, so it shouldn't screw up the filesystem or structure.
This will take me a few days to go through 18 drives ranging from 3TB to 12TB.
Once I have done that I will restart the array and let it do what its going to do, at least I will have recovered everything I could, so if it doesn't rebuild the other drive, then so be it.

I will be installing a few backup solutions going forward for my important pictures and documents, an external 12TB drive, a 4TB SSD, and my GCS account.

I will have extra protection, but better be safe than sorry again.

I do not care about my music, movies, or television files, just my docs, pictures/videos, and business files.

Alright, I'm back. That took a while, but I managed to run a recovery app on every drive and get back some of my missing data.

I have attached the diagnostics record, and a screenshot of two drives in the array, both of which are original drives, and are both 'lost' I guess.

Upon firing the server back up it did a parity check, but I have attached an image of the message where it says it has not done once since 2018.

I do not want to hit "Yes, I want to do this" and format the drive unless I have to, nor do I want to execute a "Write corrections to parity" until I understand that that will do.

This history shows a good parity was done back in May, June, and July as I have it set to do a complete parity check every month.

What are my next steps here?

I assume the 12TB drive data is just lost, but what about the 8tb? Should I execute the format, or the parity write?

Once this is all done I have a 12TB drive to add to the array.

magoosserver-diagnostics-20250815-0806.zip

Check filesystem on disk14, for the other disk, since it was formatted, the only option would be to use a file recovery app, like UFS Explorer. The free trial should show if it can recover anything.

Ok, did that, and now it no longer has the error, but is empty.

The bottom says parity is valid, but still gives me the warning it hasn't been checked since 2018, and has the Write Corrections to Parity checked.

What should I do at this point?

10 minutes ago, mrmagoo_83 said:

but still gives me the warning it hasn't been checked since 2018, and has the Write Corrections to Parity checked.

I have a vague memory that the file format of the history file was changed many years ago and there was a problem with items of the old format being mixed in with the new format that would create a problem with symptoms like you are seeing. But I can't recall the fix...

I would check that the date and time on the server is correct. Somehow it seems you managed to get some history entries written with an old date.

Date command returns: "Sat Aug 23 15:23:56 CDT 2025"

Sorry, I was traveling this past week taking my kiddo to college.

Should I run a parity check with corrections or what should be my next steps before I return to BAU?