Hello,

 

I recently had read errors on some drives. There were 8 drives impacted which led me to believe my internal SAS was the issue. I attached diagnostics to this post and have an excerpt from the log entries below when the SAS card first reported issues.

 

 

Mar 16 04:30:01 Tower root: Checking for patches for OS version 6.12.6
Mar 16 04:30:02 Tower root: Skipping 20250118234506-pr239.patch-- Already installed
Mar 17 00:33:54 Tower kernel: mpt2sas_cm1: SAS host is non-operational !!!!
### [PREVIOUS LINE REPEATED 5 TIMES] ###
Mar 17 00:33:59 Tower kernel: mpt2sas_cm1: _base_fault_reset_work: Running mpt3sas_dead_ioc thread success !!!!
Mar 17 00:33:59 Tower kernel: sd 18:0:0:0: [sdv] Synchronizing SCSI cache
Mar 17 00:33:59 Tower kernel: sd 18:0:0:0: [sdv] Synchronize Cache(10) failed: Result: hostbyte=0x01 driverbyte=DRIVER_OK
Mar 17 00:33:59 Tower kernel: sd 18:0:1:0: [sdw] Synchronizing SCSI cache
Mar 17 00:33:59 Tower kernel: sd 18:0:1:0: [sdw] Synchronize Cache(10) failed: Result: hostbyte=0x01 driverbyte=DRIVER_OK
Mar 17 00:33:59 Tower kernel: sd 18:0:2:0: [sdx] Synchronizing SCSI cache
Mar 17 00:33:59 Tower kernel: sd 18:0:2:0: [sdx] Synchronize Cache(10) failed: Result: hostbyte=0x01 driverbyte=DRIVER_OK
Mar 17 00:33:59 Tower kernel: sd 18:0:3:0: [sdy] Synchronizing SCSI cache
Mar 17 00:33:59 Tower kernel: sd 18:0:3:0: [sdy] Synchronize Cache(10) failed: Result: hostbyte=0x01 driverbyte=DRIVER_OK
Mar 17 00:33:59 Tower kernel: sd 18:0:4:0: [sdz] Synchronizing SCSI cache
Mar 17 00:33:59 Tower kernel: sd 18:0:4:0: [sdz] Synchronize Cache(10) failed: Result: hostbyte=0x01 driverbyte=DRIVER_OK
Mar 17 00:33:59 Tower kernel: sd 18:0:5:0: [sdaa] Synchronizing SCSI cache
Mar 17 00:33:59 Tower kernel: sd 18:0:5:0: [sdaa] Synchronize Cache(10) failed: Result: hostbyte=0x01 driverbyte=DRIVER_OK
Mar 17 00:33:59 Tower kernel: sd 18:0:6:0: [sdab] Synchronizing SCSI cache
Mar 17 00:33:59 Tower kernel: sd 18:0:6:0: [sdab] Synchronize Cache(10) failed: Result: hostbyte=0x01 driverbyte=DRIVER_OK
Mar 17 00:34:00 Tower kernel: sd 18:0:7:0: [sdac] Synchronizing SCSI cache
Mar 17 00:34:00 Tower kernel: sd 18:0:7:0: [sdac] Synchronize Cache(10) failed: Result: hostbyte=0x01 driverbyte=DRIVER_OK
Mar 17 00:34:00 Tower kernel: mpt2sas_cm1: mpt3sas_transport_port_remove: removed: sas_addr(0x4433221103000000)
Mar 17 00:34:00 Tower kernel: mpt2sas_cm1: removing handle(0x000a), sas_addr(0x4433221103000000)
Mar 17 00:34:00 Tower kernel: mpt2sas_cm1: enclosure logical id(0x500605b00471aaa0), slot(0)
Mar 17 00:34:00 Tower kernel: mpt2sas_cm1: mpt3sas_transport_port_remove: removed: sas_addr(0x4433221100000000)
Mar 17 00:34:00 Tower kernel: mpt2sas_cm1: removing handle(0x0009), sas_addr(0x4433221100000000)

 

I shutdown and reseated the SAS card and all SATA, SAS-to-SATA, and power cables to all drives in my array. All drives came up fine and I tried to run a parity check without selecting fix errors. I went to bed and woke up the next morning and the parity checked stopped and there were more read errors on the same drives.  I decided to turn off my server for the week until the weekend where I have time to troubleshoot.

 

I took a look today and see two drives disabled but the contents are being emulated. I shut down my server various times, reseating SATA and power cables but the two drives still show disabled. I swapped known working cables (Power and SATA) with these two drives and the other drives are fine but these two still show disabled.

 

 

The two disabled drives are healthy per checking the SMART logs. I was thinking of rebuilding the drive onto themselves after searching online but the link below is no longer valid.

 

https://wiki.unraid.net/Manual/Storage_Management#Rebuilding_a_drive_onto_itself

 

I did some more digging and found these steps. Are they correct?

 

"You can do a rebuild by doing the following: Stop array 

Unassign disk to be rebuilt 

Start array with disk unassigned 

Stop array 

Reassign disk to be rebuilt 

Start array to begin rebuild "

 

NOTE: My system has two parity drives, and I upgraded both to larger drives in the past 2-3 months. The drives that are disabled are the former parity disks. Wanted to share if that matters at all.

 

Thanks in advance.

tower-diagnostics-20250318-0750.zip

6 hours ago, Dradder1 said:

but these two still show disabled.

That's expected, as you mention after, only a rebuild can re-enable them, you can rebuild both at the same time, and the steps posted are correct, but make sure the emulated disks are mounting and contents look correct before rebuilding on top of the old disks.

Both drives are mounting so that part looks good. How would I verify the contents look correct? Would that be by starting the array as is and checking the two disabled drives contents via the Unraid gui?

 

Currently I have the system online but have not started the array. Assuming the contents are fine, I can rebuild both disks per the previously stated steps.

I started my arrary, checked the contents of the emulated disks and they look fine.

 

I stopped the array, unassigned the two disks, started array, stopped array, re-assigned, started array, and am rebuilding now.

 

Will provide an update when the rebuilding is done.

 

Thanks