I noticed today that a few of my services dropped offline, and I went to check what happened with my server. Upon first looking at the tab, it showed the contents of the disk was emulated (disk 4, I believe).

 

Because the SMART data showed as healthy, I assumed that I was all good to rebuild the array onto itself following the unraid guide of stopping the array, unallocating the disk, starting the array, stopping the array, then reallocating the disk. I noticed that the sync button did show up after I did this, but when I clicked it nothing happened, so I rebooted the server to check if that did anything.

 

After rebooting, I noticed that all of the shares in my shares tab were gone. I checked with "ls" to see that the data was actually still physically there on the disks (it is). I also tried updating to the latest version of unraid, which did make the shares show back up again, but when trying to access them over SMB it's totally empty. Additionally, all of my VMs are gone, and the docker tab says "Docker service failed to start".

Strangely, after updating and rebooting, now two of the disks show not only as disabled, but also show as unmountable.

 

I've also attached my diagnostics file. Any help is appreciated. I have all of my important data backed up on my desktop with a separate copy, so it's not the end of the world if the data is gone, but it's certainly frustrating.

 

 

tower-diagnostics-20231101-1435.zip

In addition to other issues there were write errors with disk4, replace cables and post new diags after array start.

It's in a R720xd so it's just a backplane, no cables to replace. I also don't have physical access to the server at the moment, as it's a few thousand miles away and I'm remoting in. I've tried running xfs_repair and for all of the disks it's failed with "xfs_repair: libxfs_device_zero write failed: Input/output error".

 

Here are the latest logs.tower-diagnostics-20231102-1404.zip

No point in attempting xfs_repair with disk errors, it can make things worse.