One of my array disks, disk17/sdz, seems to have dropped out of the array. 8 errors are shown in the error column.

 

The array disks are all on an LSI 9300-16e card, and on one of two disk shelves. This one is on a ds4243 with an IOM12 controller, and because the IOM12 doesn't seem to like SATA disks, is using a SAS interposer which annoyingly prevents me seeing SMART data.

 

Diagnostics attached, is anyone able to give me any sort of clue as to what has gone wrong? Not sure if I should swap the disk out or reboot and let it rebuild.

tacgnol-core-diagnostics-20240616-0843.zip

Lots of these:

 

Jun 16 07:04:22 Tacgnol-Core kernel: sd 2:0:2:0: Power-on or device reset occurred
Jun 16 07:04:22 Tacgnol-Core kernel: sd 2:0:3:0: Power-on or device reset occurred
Jun 16 07:04:22 Tacgnol-Core kernel: sd 2:0:8:0: Power-on or device reset occurred
Jun 16 07:04:22 Tacgnol-Core kernel: sd 2:0:5:0: Power-on or device reset occurred
Jun 16 07:04:22 Tacgnol-Core kernel: sd 2:0:23:0: Power-on or device reset occurred

 

Suggesting more a power/connection issue, have you tried without the interposes? It works for most users and you would get valid SMART

The Iom12 won't detect sata disks at all, I had a Dell 6gbps controller in there before which was fine, and the other shelf (which still has the dell controller) is fine. I will move the disk to the 2nd shelf when I can pull it to get smart data from it, but unbalanced is currently running to clear off some pool drives for me to swap into the array.

 

I assumed the lack of SATA support was just a limitation of the iom12 to be honest.

 

Strange for it to be a power issue, as no other drive in that shelf had an issue and the shelf has 4 power supplies! I will look to swap the drive for a 14tb spare and put it in the other shelf amd run preclear on it a couple of times.

In case anyone ever has a similar issue, I moved the disk to the IOM6 shelf and removed the interposer (so I can see the SMART data), which showed no issues.

 

Pulling one drive from the ds4243 apparently made the fans ramp down quite a bit so I've left an empty slot in that. Rebooted server and it's gone through a rebuild on that drive without issue.