Hello! 

I was greeted today by two drives (out of five) giving read errors. One is a second parity, one is a regular drive. (I do have a handful of crap unassigned devices in there as well.) 

I grabbed the diagnostics, then rebooted, I'm currently running extended smart tests on all drives. After the reboot, unraid reported all errors cleared, but I have two disabled drives now. 

I can't see any errors in the smart diags, but I reading badly on my phone until I can get home to the computer itself. 

I'd appreciate any "next step" type advice, and thoughts on trying to keep using these drives) 

 

Thanks in advance

chunksunraid-diagnostics-20201225-1046.zip

It looks as if parity, disk1, disk2 all dropped offline at the same time so it is a good chance whatever caused it was external to the drives.  If you post new diagnostics taken after the reboot they will include updated SMART reports.

Here's a fresh diag, but the extended tests haven't finished yet. Still waiting on those... 

Thanks so much for having a look. I appreciate it a lot!! 

After.Reboot.chunksunraid-diagnostics-20201225-1935.zip

Last diag update, ran this on the same reboot as the previous one, but after the extended tests had finished. I'm not even sure if the files will include these updates, but at least none of them reported errors. 

After.Long.Smart.chunksunraid-diagnostics-20201226-0701.zip

This was a problem with the onboard SATA controller:

Dec 25 10:21:38 ChunksUnraid kernel: ahci 0000:02:00.1: Event logged [IO_PAGE_FAULT domain=0x0000 address=0x00000000b4b3e000 flags=0x0000]

 

Unfortunately this issue is quite common with Ryzen boards, BIOS update might help, as well as updating to the latest Unraid beta, or disabling IOMMU if not needed.

Aha, thank you. That makes me feel a bit better, especially if it's so rare that this is the first time in a year and a half.

I'm not sure what I'd use IOMMU for, but I feel like my VM's may use it.. but I'll definitely look into making any updates that I can.