Hello, I've got errors that I can't resolve on my Cache pool (2xSSD Raid1).

Errors from server's console - attached screenshots

When I've tried to check filesystem placing pool in Maintenance Mod, I've got a lot of such messages:

checksum verify failed on 4141940736 found 000000E6 wanted FFFFFFB5
checksum verify failed on 4141940736 found 000000E6 wanted FFFFFFB5
checksum verify failed on 4141940736 found 000000E6 wanted FFFFFFB5
Csum didn't match
owner ref check failed [4141940736 16384]
ref mismatch on [46354452480 12288] extent item 1, found 0
incorrect local backref count on 46354452480 root 5 owner 304198 offset 0 found 0 wanted 1 back 0x2776150
backref disk bytenr does not match extent record, bytenr=46354452480, ref bytenr=0
backpointer mismatch on [46354452480 12288]
owner ref check failed [46354452480 12288]
ref mismatch on [46458929152 32768] extent item 1, found 0
incorrect local backref count on 46458929152 root 5 owner 304202 offset 0 found 0 wanted 1 back 0x2134640
backref disk bytenr does not match extent record, bytenr=46458929152, ref bytenr=0
backpointer mismatch on [46458929152 32768]
owner ref check failed [46458929152 32768]
ref mismatch on [46459015168 53248] extent item 1, found 0
incorrect local backref count on 46459015168 root 5 owner 304213 offset 0 found 0 wanted 1 back 0x21348a0
backref disk bytenr does not match extent record, bytenr=46459015168, ref bytenr=0
backpointer mismatch on [46459015168 53248]
owner ref check failed [46459015168 53248]
ref mismatch on [46459699200 53248] extent item 1, found 0
incorrect local backref count on 46459699200 root 5 owner 304215 offset 0 found 0 wanted 1 back 0x21356e0
backref disk bytenr does not match extent record, bytenr=46459699200, ref bytenr=0
backpointer mismatch on [46459699200 53248]
owner ref check failed [46459699200 53248]

Please suggest something to fix these errors.

Diagnostics

sirius-diagnostics-20210530-2230.zip

Tried to use command btrfs check --repair /dev/sdk1

got a lot of messages like repairing missing dir index item for inode 1137979

then

Trying to rebuild inode:304201
Moving file 'chapter4.jpg' to 'lost+found' dir since it has no valid backref
and i hope right now everything will be fine because 

btrfs scrub status: 

Error summary: no errors found

 

PS If I should do anything else please write down, thanks

 

There's uncorrectable metadata corruption on the pool, you need to backup and re-format it, there are some recovery options here if needed.

Drives fail. I thought I was smart having 2x500Gb SSD for Cache. I think I'll go 1x500Gb for cache and 500Gb backup after reading this.

 

6.