I have had two drives (2 and 4) both drop pretty frequently from my unraid server. It seems like especially if I do a parity check it will drop 4 for sure and 2 sometimes. I can rebuild with different drives, and using different connections which leads me to believe it's not an issue with my LSI controller, or my motherboard as I've used each. I'm a little confused as to where to go from here.

prometheus-diagnostics-20240318-1057.zip prometheus-syslog-20240318-1642.zip

Constant ATA errors from disk3, replace cables and post new diags.

I've replaced the cable for disk 3 and now I'm getting read errors on the disk. I set it to rebuild 3 (and 4 since it was knocked offline at the same time). It seems it didn't make it far into rebuilding at all. Here are the new diags.

prometheus-diagnostics-20240321-2114.zip

Still errors on disks 3 and 4, make sure you also check/replace the power cable, any splitters in use?

There are, need them replaced?

2 minutes ago, squazify said:

There are, need them replaced?

What type of splitter?    SATA->SATA seem to be unreliable if they split more than 2 way because the SATA connection can become current limited causing voltage sag.  Molex->SATA seem to be able to handle a 4-way split reliably.

It's Molex to Molex, then Molex to Sata. I have enough spare Molex to at least remove that first splitter.

I think I have it figured out now, thank you. Just removing a single splitter seemed to have fixed the issue, but I don't want to mark it as solved quite until I've done more testing. Thank you so much! I think at this point I'm looking into getting a new PSU that has more molex handoffs for my drives. Once again, thank you.