Constant errors since drive replacement

[Fpsware]

I recently experienced 2 faulty drives, the first was rebuild successfully, the second was in the process but appears I'll loose this - no major loss. 

 

But right now I'm experiencing a significant number of errors that are preventing basic use. 

* Dockers are failing to run/load

* Intermittently I can't write to ANY shares (zero free space)

 

The most frequent error appear in the log is...

Nov 19 17:46:44 Tower kernel: BTRFS info (device md2): the free space cache file (133164957696) is invalid, skip it
Nov 19 17:47:18 Tower kernel: BTRFS error (device md2): csum mismatch on free space cache
Nov 19 17:47:18 Tower kernel: BTRFS warning (device md2): failed to load free space cache for block group 403747897344, rebuilding it now
Nov 19 17:47:19 Tower kernel: BTRFS info (device md2): the free space cache file (419854024704) is invalid, skip it
Nov 19 17:47:20 Tower kernel: BTRFS info (device md2): the free space cache file (438107635712) is invalid, skip it
Nov 19 17:47:31 Tower kernel: BTRFS info (device md2): the free space cache file (531523174400) is invalid, skip it
Nov 19 17:47:31 Tower kernel: BTRFS info (device md2): the free space cache file (532596916224) is invalid, skip it
Nov 19 17:47:50 Tower kernel: BTRFS info (device md2): the free space cache file (751640248320) is invalid, skip it
Nov 19 17:47:50 Tower kernel: BTRFS info (device md2): the free space cache file (752713990144) is invalid, skip it
Nov 19 17:47:54 Tower kernel: BTRFS error (device md2): bad tree block start, want 943587328 have 0
Nov 19 17:48:22 Tower kernel: BTRFS error (device md2): bad tree block start, want 943702016 have 0
Nov 19 17:48:22 Tower kernel: BTRFS error (device md2): error loading props for ino 1145 (root 1): -5
Nov 19 17:48:22 Tower kernel: BTRFS error (device md2): bad tree block start, want 943702016 have 0
Nov 19 17:48:22 Tower kernel: BTRFS error (device md2): bad tree block start, want 943702016 have 0

 

Nov 19 17:54:55 Tower kernel: BTRFS info (device md1): no csum found for inode 4382249 start 146083840
Nov 19 17:54:55 Tower kernel: BTRFS warning (device md1): csum failed root 5 ino 4382249 off 146083840 csum 0xdafe0910 expected csum 0x00000000 mirror 1
Nov 19 17:55:06 Tower kernel: BTRFS warning (device md1): md1 checksum verify failed on 1024966656 wanted 2F4106F6 found 5048064A level 0
Nov 19 17:55:06 Tower kernel: BTRFS info (device md1): no csum found for inode 4382249 start 146083840
Nov 19 17:55:06 Tower kernel: BTRFS warning (device md1): csum failed root 5 ino 4382249 off 146083840 csum 0xdafe0910

tower-diagnostics-20191119-0452.zip

[Fpsware]

Running a SCRUB on Drive 1 the log fills with error messages:

 

Nov 19 17:59:29 Tower kernel: BTRFS error (device md1): bdev /dev/md1 errs: wr 2, rd 3, flush 0, corrupt 7, gen 0
Nov 19 17:59:29 Tower kernel: BTRFS error (device md1): unable to fixup (regular) error at logical 932790272 on dev /dev/md1
Nov 19 17:59:29 Tower kernel: BTRFS warning (device md1): checksum error at logical 932724736 on dev /dev/md1, physical 932724736: metadata leaf (level 0) in tree 7
Nov 19 17:59:29 Tower kernel: BTRFS warning (device md1): checksum error at logical 932724736 on dev /dev/md1, physical 932724736: metadata leaf (level 0) in tree 7

[JorgeB]

Filesystem is corrupt, best to backup any data and reformat.

[Fpsware]

So a complete reinstall of Unraid?

[JorgeB]

No, format disk1.

[Fpsware]

Disk 1 is only recently rebuilt, less than 24 hours ago. 

[JorgeB]

1 hour ago, johnnie.black said:

Filesystem is corrupt