Hello,

I found out looking at my log that one of my drive is making an error, but the error make no sense.

oct 12 16:12:21 ServRaid kernel: BTRFS error (device sdj1 state EA): level verify failed on logical 3598354939904 mirror 2 wanted 1 found 0
Oct 12 16:12:21 ServRaid kernel: BTRFS error (device sdj1 state EA): level verify failed on logical 3598354939904 mirror 1 wanted 1 found 0
Oct 12 16:12:21 ServRaid kernel: BTRFS error (device sdj1 state EA): level verify failed on logical 3598354939904 mirror 2 wanted 1 found 0
Oct 12 16:12:21 ServRaid kernel: BTRFS error (device sdj1 state EA): level verify failed on logical 3598354939904 mirror 1 wanted 1 found 0
Oct 12 16:12:21 ServRaid kernel: BTRFS error (device sdj1 state EA): level verify failed on logical 3598354939904 mirror 2 wanted 1 found 0
Oct 12 16:12:21 ServRaid kernel: BTRFS error (device sdj1 state EA): level verify failed on logical 3598354939904 mirror 1 wanted 1 found 0
Oct 12 16:12:21 ServRaid kernel: BTRFS error (device sdj1 state EA): level verify failed on logical 3598354939904 mirror 2 wanted 1 found 0
Oct 12 16:12:21 ServRaid kernel: BTRFS error (device sdj1 state EA): level verify failed on logical 3598354939904 mirror 1 wanted 1 found 0
Oct 12 16:12:21 ServRaid kernel: BTRFS error (device sdj1 state EA): level verify failed on logical 3598354939904 mirror 2 wanted 1 found 0
Oct 12 16:12:21 ServRaid kernel: BTRFS error (device sdj1 state EA): level verify failed on logical 3598354939904 mirror 1 wanted 1 found 0
Oct 12 16:12:21 ServRaid kernel: BTRFS error (device sdj1 state EA): level verify failed on logical 3598354939904 mirror 2 wanted 1 found 0
Oct 12 16:12:21 ServRaid kernel: BTRFS error (device sdj1 state EA): level verify failed on logical 3598354939904 mirror 1 wanted 1 found 0
Oct 12 16:12:21 ServRaid kernel: BTRFS error (device sdj1 state EA): level verify failed on logical 3598354939904 mirror 2 wanted 1 found 0

It talk about mirror, but it's a single drive pool

Last time that happened, I just stopped my array and restarted. Is my drive failing? Should I use xfs instead of btrfs for a single pool drive (I do plan on having more disk in the futur for that pool but not in mirror).

thank you

servraid-diagnostics-20251012-1615.zip

12 hours ago, Nodiaque said:

It talk about mirror, but it's a single drive pool

Metadata is duplicated, and there's where the issue are, both metadata copies are incorrect, and since they are 0 it suggests some metadata that should not have been deleted was.

ok, how do i fix this? Cause the disk cannot be used until I stop the array and start it back, I tried scrub, btrfs repair and it solved it for maybe 3 hours. I did a short and long smart without error.

Edited October 13, 2025Oct 13 by Nodiaque

If the other options are not working, backup and reformat may be the only way.

what other option?

There's not AFAIK, metadata where btrfs is expecting to find something.

17 hours ago, Nodiaque said:

wanted 1

Doesn't have anything:

17 hours ago, Nodiaque said:

found 0

And both metadata copies are the same, suggesting a controller or drive firmware issue incorrectly wiped that.

Ok, but this drop the drive. It stays up but can't read or write anymore. My vm that use it says drive offline and the share show empty drive. I have to stop and start the array. Happened twice in 24h

There are some btrfs recovery options below, but they don't always work with that kind of metadata corruption:

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/#findComment-543490

Oh I don't really mind losing the data. I tried all recovery and what's weird is that when it get back up, it's working. It just failed again today, was fine since monday. I also changed the sata cable and port just in case. I guess it's format and restart?

If you don't mind the data, you can just format it.

Ok. since it's single drive, should I out it on xfs? I put btrfs cause I though maybe adding another drive in the future to that pool, but it won't be needed to have these in mirror or raid.

Probably best to use xfs in that case

ok. Since I can mount it after an array stop, I'll copy the content into the array and format it. Whatever corrupted file there is isn't a big problem. It's a security camera disk and currently, there's nothing of value in it

Good evening, I've found same error in my server.
I'm not sure if these cases are related, but the difference is that I use 3 drives instead of 1.
Could you please help me with this error?

the-beast-diagnostics-20251228-1720.zip

Scrub the pool and post the results from the GUI.

P.S. the other pool is missing a device. Is that expected?

Sorry for late reply,

After I run scrub nothing happens. GUI shows "aborted" and duration 00:00:00

I don't see any missing drives. Maybe I haven't finish my hdd upgrade correctly in the past...
Were and how I can see that error ?

Thank you

the-beast-diagnostics-20260123-1953.zip

No longer remember what the issue was, but if the scrub is aborting, it means the filesystem has issues; recommend backing up and reformatting it.

Can I trust that data ? I've noticed some of my movies skips or behaves in strange ways (soundtrack changes for couple of minutes)

Also, what's the best strategy to avoid that kind of issue? Use a parity disk ? Not to use this kind of hardware ?

Should I continue to use BTRFS?

Any data that you can copy should be OK, btrfs should generate an I/O error if corruption is detected. Make sure you copy from the disk path, since FUSE in the middle can sometimes let Btrfs corruption errors through.

In my experience, Btrfs mirrors are reliable if the hardware is OK, bad hardware can cause strange issues. You can also try zfs since it now supported; it has similar features, not as flexible, but is more robust.

Thank you very much :)

After a server restart pool started showing free space (before there were none, even though used space was lower than pool total capacity) and scrub finished with no errors...

What changed - I've deleted a lot of unfinished and not started torrents (space allocation is turned off) and restarted the server.

Diagnostics atached bellow.

Can I use this pool data? And how to secure this pool with parity disk?

the-beast-diagnostics-20260127-2114.zip

Edited January 27Jan 27 by JunkoZane

Which pool do you mean? Cache is still missing a device; the other pool is raid0, so no redundancy.

Cache has all deviced (2) that it suppose to have. How can I remove this error?

I was talking about "Media" pool (correct it is raid0, but my plan is to assign one more (parity) drive to it

To correct the cache pool, you can use the CLI to remove the missing device, with the array running type.

btrfs dev remove missing /mnt/cache

Once that is done, stop the array and reimport the pool with the single device (sdf):

on main click on the first device for that pool and then "remove pool"

back on main, create a new pool with the same name and 1 slot

assign the pool device, leave the filesystem set to auto

start the array to import the pool

After that is done, you can add the other device to create a mirror.

For the other pool, you would need to add another 4TB drive first, then balance to raid1.