November 12, 20232 yr I have 2 samsung 990 pro nvme that are a few weeks old in a mirror zfs pool. Now one of them is degraded The problem started with this log Unraid kernel: nvme nvme2: I/O 249 (I/O Cmd) QID 1 timeout, aborting You can see everything in the diagnostics. Is really the HW broken? is there a way to recover it if it's not a hw problem? In scrutiny both drives seems ok unraid-diagnostics-20231112-0959.zip Edited November 12, 20232 yr by L0rdRaiden
November 12, 20232 yr Community Expert It looks like the second SSD in that pool dropped offline as indicated in your screenshot which is saying it is missing as you started getting I/O errors on it and it does not show up in the SMART reports. You might want to consider power cycling the server to see if that gets it back. It could also be worth checking that it is properly seated. There is always the chance it really has failed.
November 12, 20232 yr Author After a cold boot is working again, I did a scrub and ended ok just in case. But what is the meeaning of CKSUM at 3? do I have to do something else? do I have to be worried about this?
January 7, 20251 yr Hi @L0rdRaiden - I've got the exact same issue: a pair of 4tb 990 Pro NVMes in a zpool. One has just disappeared and the pool is in a degraded state. I've already had to RMA one of the drives because it was dropping out and would reappear after a power down/ restart, but with no SMART errors. Could you let me know how you resolved your issue?
January 7, 20251 yr Community Expert You should post the diags after a device drops, there are some kernel options that help in some cases.
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.