I have swapped out all my hardware onto a Xeon Gold 5118 server and initially all was great.  I bought two new 12TB disks to reduce the number of drives but these have not yet beem installed.

 

I have been using the Unbalanced plugin to move data to free up disks so I can remove them.  However something happened and now the array will not start after a reboot.  In the logs I see the message

"critical medium error, dev sdh, sector 976958016 op 0x0:(READ) flags 0x0 phys_seg 26 prio class 2" with a number of disk read errors and every two seconds they are repeated.  The disk in question is the parity disk.

 

At the same time another disk has been disabled.

 

Not wanting to loose any data I am trying to find out what my next steps are.  I would love some help!!

 

 

freedom-diagnostics-20240829-1707.zip

The other disk was already disabled at boot, and the parity errors are logged as a disk problem, run an extended SMART test on parity.

I have run the extended smart test on the parity drive and it only took a few seconds, it reported "Completed read failure"

freedom-smart-20240829-1946.zip

Disk2 also looks like it really failed, so some that loss may be inevitable, you can try mounting old old disk2 with UD and see what you can copy, or clone it with ddrescue.

I used UD and got all my data off Disk2 so no data loss there.

 

It sounds llike I need to replace the parity and Disk2 drives.  What is the best order to do this?

You can just do a new config with a new disk2 and new parity and re-sync parity with the remaining disks.