isp errors with QLogic 2312 / DL380 / MSA 1500

[Brad Miele]

Hello,

I have recently moved from individual HP MS30 SCSI shelves to the HP MSA 
1500 SAN with two MS30 SCSI shelves and two MS20 SATA shelves.

I am seeing the following in messages:

Dec  9 13:42:16 bwayipn04 kernel: isp0: command timed out for 0.125.4
Dec  9 13:49:58 bwayipn04 last message repeated 4 times
Dec  9 13:55:48 bwayipn04 last message repeated 2 times
Dec  9 13:59:44 bwayipn04 kernel: (da3:isp0:0:125:4): READ(10). CDB: 28 0 
6a 54 85 bf 0 0 80 0
Dec  9 13:59:44 bwayipn04 kernel: (da3:isp0:0:125:4): CAM Status: SCSI 
Status Error
Dec  9 13:59:44 bwayipn04 kernel: (da3:isp0:0:125:4): SCSI Status: Check 
Condition
Dec  9 13:59:44 bwayipn04 kernel: (da3:isp0:0:125:4): UNIT ATTENTION 
asc:29,0
Dec  9 13:59:44 bwayipn04 kernel: (da3:isp0:0:125:4): Power on, reset, or 
bus device reset occurred
Dec  9 13:59:44 bwayipn04 kernel: (da3:isp0:0:125:4): Retrying Command 
(per Sense Data)
Dec  9 13:59:45 bwayipn04 kernel: (da1:isp0:0:125:2): WRITE(10). CDB: 2a 0 
88 7a 71 9f 0 0 20 0
Dec  9 13:59:45 bwayipn04 kernel: (da1:isp0:0:125:2): CAM Status: SCSI 
Status Error
Dec  9 13:59:45 bwayipn04 kernel: (da1:isp0:0:125:2): SCSI Status: Check 
Condition
Dec  9 13:59:45 bwayipn04 kernel: (da1:isp0:0:125:2): UNIT ATTENTION 
asc:29,0
Dec  9 13:59:45 bwayipn04 kernel: (da1:isp0:0:125:2): Power on, reset, or 
bus device reset occurred
Dec  9 13:59:45 bwayipn04 kernel: (da1:isp0:0:125:2): Retrying Command 
(per Sense Data)
Dec  9 14:00:24 bwayipn04 kernel: (da2:isp0:0:125:3): READ(10). CDB: 28 0 
1 21 13 ff 0 0 20 0
Dec  9 14:00:24 bwayipn04 kernel: (da2:isp0:0:125:3): CAM Status: SCSI 
Status Error
Dec  9 14:00:24 bwayipn04 kernel: (da2:isp0:0:125:3): SCSI Status: Check 
Condition
Dec  9 14:00:24 bwayipn04 kernel: (da2:isp0:0:125:3): UNIT ATTENTION 
asc:29,0
Dec  9 14:00:24 bwayipn04 kernel: (da2:isp0:0:125:3): Power on, reset, or 
bus device reset occurred
Dec  9 14:00:24 bwayipn04 kernel: (da2:isp0:0:125:3): Retrying Command 
(per Sense Data)
Dec  9 14:08:00 bwayipn04 kernel: isp0: command timed out for 0.125.4
Dec  9 14:08:00 bwayipn04 kernel: isp0: command timed out for 0.125.4

the time out errors occur more often, and while the device recovers (it 
seems), it does stop i/o while it works itself out. This is my first 
venture into the fiber waters, and coming off of really solid performance 
with the scsi shelves i am a bit worried. can anyone offer any pointers on 
where to start looking here?

regards,

Brad
------------------------------------------------------------
  Brad Miele
  IPNStock
  bmiele at ipnstock.com