Hi guys,

My Unraid setup was running for many months without any issues. At some point my server began randomly crashing without me chaning anything at all. The only thing that I did was installing docker and Unraid updates once they became available.

Now, since a couple of days my server is barely able to run a couple of hours without crashing. Unfortunately I do not understand well enough what the diagnostics and syslog are exactly putting out to find the cause of the issue.

Can you please help me in finding out what is causing the crashes? My server is running with the latest Unraid version.

Thank you guys!

tower-diagnostics-20260323-2140.zip

Btrfs is detecting data corruption, start by running memtest.

I did a partial test with live memory tester and there have been no issues found. Might the data corruption not just be a result of too many system crashes?

58 minutes ago, skel2raw said:

Might the data corruption not just be a result of too many system crashes?

Nope, btrfs is detecting checksum errors, meaning that the data written is not the same as the data being read now.

Scrub the pool and post the results.

Indeed there seems to be a serious issue with my my btrfs cache pool (the other pool is zfs)

UUID:             a05f72c2-5645-449e-a3d9-c21a00744f10
Scrub started:    Tue Mar 24 21:32:20 2026
Status:           finished
Duration:         0:00:41
Total to scrub:   206.43GiB
Rate:             5.03GiB/s
Error summary:    csum=12
  Corrected:      0
  Uncorrectable:  12
  Unverified:     0

Pool device status shows this:

Id Path           Write errors Read errors Flush errors Corruption errors Generation errors
-- -------------- ------------ ----------- ------------ ----------------- -----------------
 1 /dev/nvme1n1p1            0           0            0               343                 0
 2 /dev/nvme2n1p1            0           0            0               343                 0

The other pool shows no errors.

Does this mean at least one drive is bad? If so, does at least one of the drives need to be replaced?

Edited March 24Mar 24 by skel2raw

Corruption is being detected on both devices. This is typically not a device problem; most often it's bad RAM, but it could also be a CPU issue.

Since there are uncorrectable errors, look in the syslog for the list of the affected files and delete/restore them from a backup, then run another scrub to confirm 0 errors, and if yes, reset the stats. After that, keep monitoring the pool, if more errors show up, there's still a problem.

7.3.0-beta will monitor the pool now; for previous releases, you can use this: https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/#findComment-700582

So I found the files and deleted them. But 2 uncorrectables remain. The syslog only shows this info and no file:

Mar 25 18:02:57 Tower kernel: BTRFS error (device nvme0n1p1): unable to fixup (regular) error at logical 55138320384 on dev /dev/nvme0n1p1 physical 50843353088
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6891, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6890, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6889, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6888, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6887, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6886, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6885, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6884, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6883, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6882, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6881, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6876, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6875, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6874, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6873, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme0n1p1, physical 50843353088, root 6722, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:02:57 Tower kernel: BTRFS error (device nvme0n1p1): bdev /dev/nvme0n1p1 errs: wr 0, rd 0, flush 0, corrupt 18, gen 0
Mar 25 18:03:06 Tower kernel: BTRFS error (device nvme0n1p1): unable to fixup (regular) error at logical 55138320384 on dev /dev/nvme2n1p1 physical 50822381568
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6891, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6890, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6889, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6888, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6887, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6886, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6885, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6884, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6883, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6882, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6881, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6876, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6875, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6874, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6873, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS warning (device nvme0n1p1): checksum error at logical 55138320384 on dev /dev/nvme2n1p1, physical 50822381568, root 6722, inode 2598, offset 0, length 4096, links 1 (path: usr/bin/psl)
Mar 25 18:03:06 Tower kernel: BTRFS error (device nvme0n1p1): bdev /dev/nvme2n1p1 errs: wr 0, rd 0, flush 0, corrupt 18, gen 0

That suggests it's in the metadata; I recommend backing up and recreating the pool.

Can I do the move from the cache pool with unbalanced or should I use the mover?

You should be able to use what you prefer, but note that some files may fail to move due to the corruption.

Okay, so I have recreated the pool. But my server crashed again after about 1 day of running. Any idea what I should do next?

tower-diagnostics-20260328-2009.zip

Edited March 28Mar 28 by skel2raw

You really should run memtest86 for several hours. If RAM has problems it will continue to create other problems.

So I have let run memtest86 pro now for an entire night and it finished the 4 rounds of tests without any errors. Any other ideas? 😅

Enable the syslog server and post that after the next crash, in case there's something more logged.

Syslog is enabled but after the last crash Unraid does not log anything at all anymore. Log is simply empty. Any idea how to fix this?

I suggest posting a screenshot of your syslog settings. It is not uncommon for them not to be correct for a server to log to itself.

Set "remote syslog server" to localhost

Still no log

Post another screenshot of syslog server settings.

I tried various settings but it still won't show up.

That looks OK to me, post the output from ls -l /mnt/user/Temp

It's empty, no files inside

Post the full diags.

4 hours ago, skel2raw said:

It's empty, no files inside

Previous diagnostics shows user share Temp exists on multiple array disks. Could be empty on all though.

Minimum Free for Temp is 188G but several disks have that much.