Hello,

 

a few weeks back I tried those unraid dashboard dockers which resulkted in full cache / nvme pretty quick. also I felt not realy confident with running them privileged and all I wanted was traffic per docker. All this overkill and issues with the cache led me to relatively quickly uninstall them again and use vnstat.

 

since then I had I think one reboot and so far no issues - everything running fine. Today I watched some stuff via plex as suddently my connection did not work anymore.

 

Log is spammed with: Dec 29 22:19:05 godzilla kernel: BTRFS error (device nvme0n1p1: state EA): parent transid verify failed on logical 1557527003136 mirror 1 wanted 1799110 found 1794281

 

Cache ist at 61GB / 1TB read only it seems.

 

root@godzilla:~# btrfs dev stats /mnt/cache
[/dev/nvme0n1p1].write_io_errs    0
[/dev/nvme0n1p1].read_io_errs     0
[/dev/nvme0n1p1].flush_io_errs    0
[/dev/nvme0n1p1].corruption_errs  0
[/dev/nvme0n1p1].generation_errs  0

 

I then tried to scrub the docker BTRFS without luck - aborted - no entry in the log. Where do I see why?

 

Then I try to stop the array and it can not stop the array because:

Dec 29 22:25:57 godzilla emhttpd: Unmounting disks...
Dec 29 22:25:57 godzilla emhttpd: shcmd (5519207): umount /mnt/cache
Dec 29 22:25:57 godzilla root: umount: /mnt/cache: target is busy.
Dec 29 22:25:57 godzilla emhttpd: shcmd (5519207): exit status: 32
Dec 29 22:25:57 godzilla emhttpd: Retry unmounting disk share(s)...

 

This is now ongoing for 25 mins + and I do not know what I should do

 

 

 

godzilla-diagnostics-20231229-2243.zip

update:

 

root@godzilla:~# umount -l /dev/loop2

root@godzilla:~# dmsetup info
No devices found

root@godzilla:~# umount -l /dev/loop2

again fixed it. 🤷‍♀️

 

Scrub started: Sat Dec 30 00:15:52 2023 Status: finished Duration: 0:00:13 Total to scrub: 7.56GiB Rate: 595.68MiB/s Error summary: no errors found

 

Everything is running again. Still, if anybody could tell me why my cache went read only that would be great! Because I do not trust it it right now

Dec 29 21:52:04 godzilla kernel: BTRFS error (device nvme0n1p1): block=1557526740992 write time tree block corruption detected

 

It went read only because of this, this error usually means a hardware issue, most often bad RAM, but with the current kernel there have been some possible false positives, so I would recommend running memtest and if nothing is found try to recreate the pool or use zfs instead.

On 12/30/2023 at 11:53 AM, JorgeB said:

Dec 29 21:52:04 godzilla kernel: BTRFS error (device nvme0n1p1): block=1557526740992 write time tree block corruption detected

 

It went read only because of this, this error usually means a hardware issue, most often bad RAM, but with the current kernel there have been some possible false positives, so I would recommend running memtest and if nothing is found try to recreate the pool or use zfs instead.

 

Thank you JorgeB. 28h MEMTest with 0 errors

and I will convert the cache to ZFS the next days. Also want to try out other hardware. I realy hope this does not happen again  it ran 1.5y pretty flaweless.

cheers!