Solved LSI SAS 9300-8i mpr issue

[Fluxor]

Hi All,

I've been scratching my head over this issue that I've been having. I've been getting the following errors when trying to write to a raidz1 array of four disks. The writes go through but I get a healthy amount of these messages. I have verified my cabling from the card to the SAS back plane. This is all new gear. Does anyone know what this is or perhaps could point me in a direction to decipher the mpr0 log info ? I have googled the issue and only had bad cables and drives turn up. Thanks in advance.

kernel: mpr0: log_info(0x31120303): originator(PL), code(0x12), sub_code(0x0303)
kernel: (da1:mpr0:0:9:0): WRITE(10). CDB: 2a 00 58 00 9b e8 00 00 48 00 length 36864 SMID 510 terminated ioc 804b scsi 0 state c xfer 0
kernel: (da1:mpr0:0:9:0): WRITE(10). CDB: 2a 00 58 00 9b e8 00 00 48 00
kernel: (da1:mpr0:0:9:0): CAM status: CCB request completed with an error
kernel: (da1:mpr0:0:9:0): Retrying command

 

[Fluxor]

I may have found the root of my problem. The drives are connected to a back plane. The back plane has 2 connectors, both of which are connected back to the LSI 9300 8i. I think that the drives were being detected on both ports. I removed one of the cables and these errors went away. All drives are still being detected and used. If anyone has a similar experience or more info, I would appreciate any light that could be shed. Thanks in advance.

 

[SirDice]

Have a look at the documentation of the backplane. The second connection might be to provide fault-tolerance and is intended to be used with a second card.