My server has been acting strange the last week and i was trying to troubleshoot it myself and have failed. I was attempting to remove a data disk and shrink my array. I removed all the data off the drive, zero'd it using the user script, removed it, and then was checking parity and during parity disk 7 had like 2000 UDMA CRC errors and was unmounted, pausing the check. I tried resuming but found instead that it just kept erroring. I decided to reboot the server and attempt to just restart the check, but after the reboot disk 7 is now unmountable with a wrong or no file system. So i started to rebuild that disk and this time disk 3 started giving millions of errors and a few CRC errors. so i rebooted again and started rebuild again, and now i have disk 5 giving millions of errors and disk 3 still ticking up CRC errors and disk 7 still unmountable.

 

From what i can tell googling this is usually a sata data cable problem, but to have so many "random" errors across disks im starting to think my PCIE SAS adapter is dying maybe? Looking at the cables, these drives are not even on the same breakout cable

 

Can anyone see anything in the diagnostics from this last boot with CRC notifications on disk 3 and errors on disk 5?

 

 

 

 

tower-diagnostics-20230308-0706.zip

Disk5 dropped offline, check/replace cables and post new diags after array start.