This morning, I realized that my cache pool (1TB drive and a 2TB drive) had filled up and the mover was not running. When I invoked the mover, it would stop almost instantly, so I turned on logging to see what was happening. I found that my 1TB cache drive had entered read-only mode. I restarted the array several times and was able to get mover to move several files off of the cache, but every time, it would go back into read-only mode. I unmounted the 1TB drive to attempt to move everything off the 2TB drive and start a rebalance. I realized after unmounting it that mover wouldn't run without the 1TB drive in the pool, so I readded it. Now it's showing as "Unmountable: No File System." 

 

Do I need to reformat both drives in the cache pool and is there anyway to recover my data before doing so? Both drives were running a brtfs file system while my main array is running xfs.

 

EDIT: Currently attempting to recover data from it by following one of the topics in the faq. Currently getting this error when trying to pull from my 2TB drive

Feb  5 13:22:25 GAMMA kernel: BTRFS info (device nvme0n1p1): enabling all of the rescue options
Feb  5 13:22:25 GAMMA kernel: BTRFS info (device nvme0n1p1): ignoring data csums
Feb  5 13:22:25 GAMMA kernel: BTRFS info (device nvme0n1p1): ignoring bad roots
Feb  5 13:22:25 GAMMA kernel: BTRFS info (device nvme0n1p1): disabling log replay at mount time
Feb  5 13:22:25 GAMMA kernel: BTRFS info (device nvme0n1p1): using free space tree
Feb  5 13:22:25 GAMMA kernel: BTRFS info (device nvme0n1p1): has skinny extents
Feb  5 13:22:25 GAMMA kernel: BTRFS error (device nvme0n1p1): devid 1 uuid 6662b2d4-19a1-4961-904c-ea10bbad500c is missing
Feb  5 13:22:25 GAMMA kernel: BTRFS error (device nvme0n1p1): failed to read chunk tree: -2
Feb  5 13:22:25 GAMMA kernel: BTRFS error (device nvme0n1p1): open_ctree failed

 

gamma-diagnostics-20230205-1243.zip

Edited February 5, 20233 yr by TaylorM

Stop the array and post the output of:

btrfs-select-super -s 1 /dev/nvme1n1p1

then

btrfs fi show