February 9, 20242 yr Started to have dockers stop running one day last week. I notice that the syslog is full of errors like this. Quote Feb 8 21:23:14 Tower kernel: BTRFS error (device nvme0n1p1): bdev /dev/nvme0n1p1 errs: wr 0, rd 0, flush 0, corrupt 3175, gen 0 Some forum post states that it might be the 6.12.x version causing issues so I downgrade to 6.11.5. I had my main docker plex come back up but not 2 other dockers give me errors when I try to launch and the BTRFS errors persist, but not as many. The short smart test passes, but shows this at the bottom. Quote Error Information (NVMe Log 0x01, 16 of 256 entries) Num ErrCount SQId CmdId Status PELoc LBA NSID VS - [7 unused entries] 7 562954730668031 - - - - 0 0 - I stopped docker and vm plugins, set my cache folders to yes, and started the mover in case its the drive. I do have spare cache drives I could use but still want to figure out what the issue is. tower-diagnostics-20240208-2144.zip
February 10, 20242 yr Author On 2/9/2024 at 1:26 AM, JorgeB said: Btrfs is detecting data corruption, start by running memtest. Mem test passed.
February 11, 20242 yr Author 3 hours ago, JorgeB said: Post the output of a correcting scrub on the pool How do I do that for the pool?
February 11, 20242 yr Author 10 minutes ago, JorgeB said: Click on the pool and scroll down to the scrub section. UUID: dd5df0a2-476a-485c-bb04-520924ae1af2 Scrub started: Sun Feb 11 10:33:10 2024 Status: finished Duration: 0:00:48 Total to scrub: 86.37GiB Rate: 1.80GiB/s Error summary: csum=5 Corrected: 0 Uncorrectable: 0 Unverified: 0
February 11, 20242 yr Author Checked the repair box and all are unrecoverable. Seems like I should format and reuse the drive. UUID: dd5df0a2-476a-485c-bb04-520924ae1af2 Scrub started: Sun Feb 11 10:34:46 2024 Status: finished Duration: 0:00:49 Total to scrub: 86.37GiB Rate: 1.76GiB/s Error summary: csum=5 Corrected: 0 Uncorrectable: 5 Unverified: 0
February 11, 20242 yr Community Expert Solution Check the syslog, it should list the corrupt files, delete/restore them from a backup, but if new corruptions keep happening there's likely still an underlying hardware issue.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.