5 Drives showing read errors with 2 total failure

[Wayne66]

This afternoon I had 5 drives start showing read errors with 2 failed. I have no idea what happened or what to do to fix. 

 

Since one of the read error drives is my parity drive and the parity 2 has failed, am I screwed and have lost everything on the drives lost?

 

tower-diagnostics-20200131-1746.zip

[JorgeB]

Controller problem, possibly related to virtualization:

Jan 31 15:27:24 Tower kernel: ahci 0000:01:00.1: Event logged [IO_PAGE_FAULT domain=0x0000 address=0x00000000fca10000 flags=0x0000]

Not the first time I see these errors with AMD based systems, rebooting will clear the disk errors, then rebuild/re-sync the disabled disks, make sure the emulated disk4 is mounting correctly before rebuilding on top of the old disk, bios update might help so it doesn't reoccur.

[Wayne66]

Thanks for the reply. I just happened to have 2 extra drives available to replace those disabled drives and it is performing a parity sync/rebuild currently just like you said. But you are saying that I could have left the disabled drives in place and when the array started back up it would have done the same for them, rebuild the data? Even if they show with "disabled"? Is the "disabled" flag not related to that specific disk, but rather to it's spot in the array?

 

Darn, I should have waited for your response, but I felt compelled to replace the disabled drives and I haven't lost any data. So I guess it's still a win. I also should be able to add those drives back into the array later for more storage, right? I expect they are just fine.

[JorgeB]

3 minutes ago, Wayne66 said:

But you are saying that I could have left the disabled drives in place and when the array started back up it would have done the same for them, rebuild the data?

Yes, you just needed to re-enable both:

https://wiki.unraid.net/Troubleshooting#Re-enable_the_drive