Update: i managed to mount the cache drive in read-only, copy the contents to an array disk, format the cache drive, and punt the backup back. See details below.

I came to my Unraid server being unrepsonsive and all docker and VMS stopped, SMB shares still working.

It's an Intel NUC with one nvme0n 512GB drive cache, and one WD Red 8TB drive for the arrya, no parity.

dmesg shows an BTRFS critical error, i don't know what to do next, help :-)

[ 171.074461] BTRFS: error (device nvme0n1p1: state A) in __btrfs_free_extent:3110: errno=-117 Filesystem corrupted

[ 171.074470] BTRFS info (device nvme0n1p1: state EA): forced readonly

[ 171.074473] BTRFS info (device nvme0n1p1: state EA): leaf 122585088 gen 470682 total ptrs 153 free space 4349 owner 2

[ 171.074656] I/O error, dev loop2, sector 0 op 0x1:(WRITE) flags 0x800 phys_seg 0 prio class 2

[ 171.074658] item 31 key (596771348480 168 4096) itemoff 14587 itemsize 53

[ 171.074660] extent refs 1 gen 391795 flags 1

[ 171.074662] ref#0: extent data backref root 5 objectid 27178 offset 13238718464 count 1

[ 171.074664] item 32 key (596771352576 168 102400) itemoff 14534 itemsize 53

[ 171.074666] extent refs 1 gen 470681 flags 1

[ 171.074665] BTRFS error (device loop2): bdev /dev/loop2 errs: wr 0, rd 0, flush 1, corrupt 0, gen 0

[ 171.074667] ref#0: extent data backref root 5 objectid 666178 offset 30353338368 count 1

[ 171.074670] item 33 key (596771454976 168 40960) itemoff 14481 itemsize 53

[ 171.074670] BTRFS warning (device loop2): chunk 13631488 missing 1 devices, max tolerance is 0 for writable mount

[ 171.074672] extent refs 1 gen 470681 flags 1

[ 171.074674] ref#0:

[ 171.074673] BTRFS: error (device loop2) in write_all_supers:4094: errno=-5 IO failure (errors while submitting device barriers.)

[ 171.074674] extent data backref root 5 objectid 666178 offset 30517137408 count 1

[ 171.074676] BTRFS info (device loop2: state E): forced readonly

[ 171.074681] item 34 key (596771495936 168 20480) itemoff 14428 itemsize 53

[ 171.074682] BTRFS error (device loop2: state EA): Transaction aborted (error -5)

[ 171.074683] extent refs 1 gen 470681 flags 1

[ 171.074685] ref#0:

[ 171.074684] BTRFS: error (device loop2: state EA) in btrfs_sync_log:3146: errno=-5 IO failure

BTRFS critical (device nvme0n1p1: state EA): unable to find ref byte nr 596772597760 parent 0 root 5 owner 666178 offset 45822738432 slot 108

[ 144.343343] BTRFS error (device nvme0n1p1: state EA): failed to run delayed ref for logical 596772597760 num_bytes 65536 type 178 action 2 ref_mod 1: -2

[ 144.343348] BTRFS: error (device nvme0n1p1: state EA) in btrfs_run_delayed_refs:2184: errno=-2 No such entry

[ 144.343359] BTRFS: error (device nvme0n1p1: state EA) in btrfs_replay_log:2084: errno=-2 No such entry (Failed to recover log tree)

[ 144.359371] BTRFS error (device nvme0n1p1: state EA): open_ctree failed

Solved:

managed to mount the cache drive read only and copy the contents to an array disk, formatted the nvme drive, and copied the contents back.

mkdir /mnt/temp

mount -o ro,relatime,skip_balance,space_cache=v2 /dev/nvme0n1p1 /mnt/temp

cp -a /mnt/temp /mnt/user/backup_nvme

edited /boot/config/disk.cfg and set startArray="no"

reboot

umount -f /dev/nvme0n1p1

mkfs.btrfs -f /dev/nvme0n1p1

edited /boot/config/disk.cfg and set startArray="yes"

reboot

mount -o ro,relatime,skip_balance,space_cache=v2 /dev/nvme0n1p1 /mnt/temp

cp -a /mnt/user/backup_nvme /mnt/temp

Edited August 4, 2025Aug 4 by peewee67

Please post the diagnostics.