On the 1st of August my scheduled parity check kicked of and maybe an hour into it a disk was dropped from the array. I replaced the disk and allowed it to rebuild which it did successfully. Today I decided to run a non-correcting parity check to make sure everything was fine and about 30 minutes into the check I have a disk drop from the array, it was an old drive but had not issues in the smart report and I should have maybe stopped there but I decided to replace it since it was plausible that it had failed. About 2 hours into the rebuild another drive dropped. Obviously something else is wrong. I'm attaching a diagnostic zip file that should include today's failure and the rebuild I started and then the additional failure. At this point I am running memtest86+ but so far no errors. The system has 3 LSI controllers in it and I'm starting to wonder if it's not one of those. Any help would be appreciated.

Diags.zip

Looks more like a power/connection problem.

It could be either of those but it's an EVGA 1600 watt PSU and a 24 bay SAS backplane, I can find another PSU but not sure what I can do about the backplane other than to make sure the SFF cables are well seated but since those are have clips they are probably seated well. It is likely very dusty on the inside, is that a possibility (maybe a controller or somgetting else getting hot due to dust)?

 

Also, I did stop the rebuild of disk 5 once disk 6 was dropped, when I boot it back up will disk 6 have the error condition reset and can I restart the rebuild of disk 5 without having to do a new config?

Since you have dual parity, and assuming both emulated disks are mounting, it's better to rebuild both

Currently I have drive 5 rebuilding. Disk disk is currently disabled/emulated during the rebuild. I don't believe there is anything wrong with disk 6, when disk 5 is done rebuilding how do i get disk6 to rebuild? Presumably the disabled status needs to be reset.

https://docs.unraid.net/unraid-os/manual/storage-management#rebuilding-a-drive-onto-itself