I built a rig and am trying out Unraid.  Here's the specs:

Gigabyte B365M DS3H Board

Intel i3-8100 Proc

16G G.Skill

9210-8i SAS in IT MODE

Qty 4 Seagate 6TB SAS Disks (3 Array, 1 Parity)

Kingson 1TB SSD (Cache)

Roswiel 850W 80Plus Bronze PSU

 

When I boot from my Stick, Unraid comes up fine.  I setup my Array with 1 - 6TB Parity, 3 - 6TB Array Members, 1 - 1TB SSD for Cache.  As soon as I attempt to start the array, I get 384 Errors on the parity disk and the Read Check Pauses itself.

 

I've tried replacing my SAS cable, tried changing the order of the disks, tried using just two disks, tried removing the SSD.... Can't figure out why I keep getting these parity errors ONLY on the party disk, even though the Parity disk has been swapped out a few times with other disks (validated by Serial number).  Not sure what else could be going on... advice??

 

Go to Tools-diagnostics and attach the complete Diagnostics zip file to your NEXT  post 

Attached.  I'm suspecting there might be an issue with my HBA....

tower-diagnostics-20191206-1348.zip

I tried reflashing the IT mode Firmware and Bios settings on the HBA.  Same result though.  bombed while trying to format.  latest diag logs attached.  

tower-diagnostics-20191206-2253.zip

OK now I'm even more stumped.  I happen to have an LSI SAS2116 PCI-Express Fusion-MPT SAS-2 [Meteor] laying around and threw that in here.  Took the drives and connected it to the new card, same issue.... So I either have a config issue or maybe some other hardware.  One thing I noticed, no bios for either SAS card appears at boot.  AMI bios, right to Unraid Boot... I might give FreeNas a try to see if it can make something out of these disks to see if there might be some other hardware related issue.  Any help that you mods can provide is appreciated! 

It should be disk problem, all write error and have many "Errors Corrected by ECC"

 

Error counter log:
           Errors Corrected by           Total   Correction     Gigabytes    Total
               ECC          rereads/    errors   algorithm      processed    uncorrected
           fast | delayed   rewrites  corrected  invocations   [10^9 bytes]  errors
read:    8153795        0         0   8153795          0       2060.176           0
write:         0        0         0         0          0        276.377           0
verify:     7991        0         0      7991          0          0.000           0
 

 

Those drive seems logical 512 bytes, working before ?

May be try some kind of low level format, but I don't know this action correct or not.

I saw that but I've swapped out the drive with another and had the exact same problem.  The Party drive appears to always generate tons of errors.  I also tried removing the party drive completely and just using two disks in the array.  No errors when creating the array but it generated 32 errors when I start a format and then it aborts.... I'm going to see if I can get the drives to function with LVM or maybe try FreeNas

 

 

To me looks more like a disk problem, not necessarily bad disks but something related to them, do you have a different model disk you could try, maybe even a couple of old SATA disks just to test?

OK I'm lost.  I did a complete disk check, 0 errors.  Tried to format.  Errorred out.  Smart reports errors...but wouldn't those errors cause the check to fail as well, or at least report the errors?  Never had an issue like this with disks before.  GEtting replacement disks to try.... I'm at a loss...

tower-diagnostics-20191218-1900.zip

It's like the disks don't allow any writes, why I suspect it's something related to them.