Question at the bottom.
I would like to use a Sun StorageTek 6100 as JBOD for ZFS, and so I have connected it to OmniOS over FC with a QLC2462 HBA. The disk array doesn't have a controller tray, just an expansion tray, so no raid controller.
Both ports on the HBA are connected to the disk array in port 1B, so I am going for multiplexing. For one of the links I get errors on all the disks, where the other link is perfect.
# fcinfo remote-port -sl -p 210000e08b89bd75
Remote Port WWN: 200100a0b84f89cf
Active FC4 Types:
SCSI Target: yes
Port Symbolic Name:
Node WWN: 200000a0b84f89cf
Link Error Statistics:
Link Failure Count: 0
Loss of Sync Count: 5
Loss of Signal Count: 8
Primitive Seq Protocol Error Count: 0
Invalid Tx Word Count: 76
Invalid CRC Count: 0
LUN: 0
Vendor: ATA
Product: HITACHI HUA7210S
OS Device Name: /dev/rdsk/c3t200000A0B84F89CFd0s2
...
I have tried to replace the fibre, but the errors persist, and don't increase.
When looking at format
I see 4 of the disks as expected, and not the remaining 10.
...
/scsi_vhci/disk@g5000c5006443582b
36. c3t200000A0B84EC566d0 <ATA-HITACHIHUA7210S-0609 cyl 60798 alt 2 hd 255 sec 126>
/scsi_vhci/disk@g200000a0b84ec566
37. c3t200000A0B84F2D06d0 <ATA-HITACHI HUA7210S-0609-931.51GB>
/scsi_vhci/disk@g200000a0b84f2d06
...
cfgadm -al
shows all 14 disks for both links.
Question
When there is no controller tray in the disk array, why am I not seeing all the disks?
Can the link error have something to do with this? If, how can it be fixed or reset?