Looking for general advice or a sanity check.

During a scheduled parity check, one of my two parity drives plus drive 1 of the array became "Disabled." Weird but ok...I take the array offline and run extended SMART tests on both drives, they both pass. I remove them both from the array, add them back, and start the array again, which automatically starts to rebuild the array.

 

During the rebuild, I notice disk 10 is returning millions of errors, nearly as many as the writes. I stop the array again, run extended SMART on disk 10, which also passes. I start to rebuild again.

Disk 10 still throws errors, but now disks 3 and 8 are also throwing errors. The rebuild fails, obviously. And just as a cherry on top...the crappy old 2TB drive I was using as a cache for torrent downloads is also showing a SMART error.

 

Five stage of grief later, and I've come to grips with all my data being gone. Luckily it wasn't anything I can't just download again with enough time and patience. I'm more interested to know how this could have happened, especially with no prior warning?? I have notifications on, they all go to my private Discord server and I get them instantly on my phone, and there was no prior warning for any of this. I'm not using a UPS, could this have been a power event? Do I need to replace ALL the drives with errors, or can I pre-clear them and start a new array if they pass extended SMART?

Cheers,
--unraid-diagnostics-20230703-0603.zip

Log only shows the first issue with disk10, that disk dropped offline, SMART looks good, looks more like a power/connection issue, replace cables and try again.