I've been getting errors with my cache drive since i upgraded to a ryzen 3900x and the x570 aorus pro. i have two 500gb nvme drives in the cache pool:

 

Quote

Oct 17 18:55:06 Tower kernel: BTRFS error (device nvme0n1p1): block=170529144832 write time tree block corruption detected
Oct 17 18:55:06 Tower kernel: BTRFS: error (device nvme0n1p1) in btrfs_commit_transaction:2327: errno=-5 IO failure (Error while writing out transaction)
Oct 17 18:55:06 Tower kernel: BTRFS info (device nvme0n1p1): forced readonly
Oct 17 18:55:06 Tower kernel: BTRFS warning (device nvme0n1p1): Skipping commit of aborted transaction.
Oct 17 18:55:06 Tower kernel: BTRFS: error (device nvme0n1p1) in cleanup_transaction:1898: errno=-5 IO failure

i have these issues on the latest stable and the new beta 30 unraid

 

tower-diagnostics-20201017-2258.zip

Edited October 18, 20205 yr by mlapaglia

well i just ran a memtest and got 240 errors within a minute, so it's probably that.

EDIT: 1x16GB stick out of 4 was bad, luckily hyperx has a lifetime warranty

Edited October 18, 20205 yr by mlapaglia

6 hours ago, mlapaglia said:

write time tree block corruption detected

Yep, this is usually a indication of bad RAM.