I've run into a strange issue where one of my slots in my JBOD is unrecognizable in UNRAID and shows "Device has an error status: CANT_OPEN". I say one of the slots because I've tried swapping the drives around and it's always that slot. I installed Windows on my server on a spare drive in another slot and confirmed that the drives shows up and is read/writable (just created a simple volume and copied some files to it). What could be causing this?

Jul 11 17:40:36 KRZ-UR kernel: sd 0:0:10:0: [sdk] 7814037168 512-byte logical blocks: (4.00 TB/3.64 TiB)
Jul 11 17:40:36 KRZ-UR kernel: sd 0:0:10:0: [sdk] Write Protect is off
Jul 11 17:40:36 KRZ-UR kernel: sd 0:0:10:0: [sdk] Mode Sense: d7 00 10 08
Jul 11 17:40:36 KRZ-UR kernel: sd 0:0:10:0: [sdk] Write cache: disabled, read cache: enabled, supports DPO and FUA
Jul 11 17:40:36 KRZ-UR kernel: sd 0:0:10:0: [sdk] Preferred minimum I/O size 512 bytes
Jul 11 17:40:36 KRZ-UR kernel: sdk: sdk1
Jul 11 17:40:36 KRZ-UR kernel: sd 0:0:10:0: [sdk] Attached SCSI disk
Jul 11 17:42:46 KRZ-UR emhttpd: online: ST4000NM0025_ZC14W4NL_35000c5009500a0ef (sdk) 512 7814037168 => data partition: sdk1 type: gpt start: 2048 size: 3907016704 code: 0xee (8)
Jul 11 17:42:48 KRZ-UR emhttpd: import 71 pool device: (sdk) ST4000NM0025_ZC14W4NL_35000c5009500a0ef
Jul 11 17:42:48 KRZ-UR emhttpd: read SMART /dev/sdk
Jul 11 17:45:56 KRZ-UR emhttpd: import 71 pool device: (sdk) ST4000NM0025_ZC14W4NL_35000c5009500a0ef
Jul 11 17:46:01 KRZ-UR emhttpd: shcmd (120): /usr/sbin/zpool import -f -m -N -o autoexpand=on  -d /dev/sdx1 -d /dev/sdo1 -d /dev/sdt1 -d /dev/sdw1 -d /dev/sdv1 -d /dev/sda1 -d /dev/sdm1 -d /dev/sdu1 -d /dev/sdj1 -d /dev/sdn1 -d /dev/sdq1 -d /dev/sdg1 -d /dev/sdr1 -d /dev/sdd1 -d /dev/sdi1 -d /dev/sde1 -d /dev/sds1 -d /dev/sdh1 -d /dev/sdf1 -d /dev/sdc1 -d /dev/sdp1 -d /dev/sdl1 -d /dev/sdb1 -d /dev/sdk1 1528848933852571142 zfs
Jul 11 17:46:09 KRZ-UR emhttpd:               16524075120955047720  OFFLINE      0     0     0  was /dev/sdk1/old
Jul 11 17:46:09 KRZ-UR emhttpd:               5541448863137311936   UNAVAIL      0     0     0  was /dev/sdk1
Jul 11 17:46:11 KRZ-UR emhttpd: writing GPT on device (sdk) with data partition offset 64
Jul 11 17:46:11 KRZ-UR emhttpd: shcmd (126): wipefs -af --lock /dev/sdk
Jul 11 17:46:11 KRZ-UR root: /dev/sdk: 8 bytes were erased at offset 0x00000200 (gpt): 45 46 49 20 50 41 52 54
Jul 11 17:46:11 KRZ-UR root: /dev/sdk: 8 bytes were erased at offset 0x3a3817d5e00 (gpt): 45 46 49 20 50 41 52 54
Jul 11 17:46:11 KRZ-UR root: /dev/sdk: 2 bytes were erased at offset 0x000001fe (PMBR): 55 aa
Jul 11 17:46:11 KRZ-UR emhttpd: shcmd (127): sgdisk -o -a 8 -n 1:32K:0 /dev/sdk
Jul 11 17:46:11 KRZ-UR kernel: sdk: sdk1
Jul 11 17:46:12 KRZ-UR kernel: sdk: sdk1
Jul 11 17:46:12 KRZ-UR emhttpd: shcmd (128): partprobe /dev/sdk
Jul 11 17:46:12 KRZ-UR kernel: sdk: sdk1
Jul 11 17:46:20 KRZ-UR emhttpd: shcmd (131): /sbin/wipefs -af --lock /dev/sdk1
Jul 11 17:46:20 KRZ-UR emhttpd: shcmd (132): /usr/sbin/zpool add -f zfs  raidz2 /dev/sdk1 
Jul 11 18:01:23 KRZ-UR emhttpd: spinning down /dev/sdk
Jul 11 18:01:23 KRZ-UR SAS Assist v2024.11.25: Spinning down device /dev/sdk
Jul 11 18:01:36 KRZ-UR emhttpd: read SMART /dev/sdk

The "can't open" error is just a visual issue, and it can be ignored; it resolves itself once the rebuild is done. https://product.unraid.net/p/7-3-1-zfs-pool-can-incorrectly-show-one-or-more-online-devices-as-cant_open-during-a-resilver-display-issue-only

You do now have a different problem, you will need to use the CLI to remove one of the "replacing" devices, but to see the current pool status, please post the diagnostics first.

5 hours ago, JorgeB said:

The "can't open" error is just a visual issue, and it can be ignored; it resolves itself once the rebuild is done. https://product.unraid.net/p/7-3-1-zfs-pool-can-incorrectly-show-one-or-more-online-devices-as-cant_open-during-a-resilver-display-issue-only

You do now have a different problem, you will need to use the CLI to remove one of the "replacing" devices, but to see the current pool status, please post the diagnostics first.

Yes, I actually fixed this last night by detaching the non-existent drives so as to replace the one that died.

zpool clear zfs

zpool detach zfs 5541448863137311936

zpool replace -f zfs 16524075120955047720 /dev/sdk

One curious thing after the resilvering process was completed is that my replacement drive still shows up as "New Device" with a blue square rather than a green circle. I've rebooted the server and the array didn't start automatically as it said that the drive would be overwritten again when it started. I started it and it didn't overwrite and the zfs pool looks fine... Is this a bug?

11 hours ago, Kreeze said:

I started it and it didn't overwrite and the zfs pool looks fine... Is this a bug?

That can happen when you use the CLI, and the GUI doesn't match the pool status; you can resolve it by reimporting the pool

on main click on the first device for that pool and then "remove pool"

back on main, create a new pool with the same name and number of slots

assign the pool device(s), leave the filesystem set to auto

start the array to import the pool

You can also do a new config to avoid reassigning all the devices:

-Tools -> New Config -> Preserve current assignments: All -> Apply