3ware 8006-2LP on Linux 2.6 drive error, seagate disks

3ware 8006-2LP on Linux 2.6 drive error, seagate disks

[Marcin Hlybin]

Hello,

Recently I've come across a lot of problems with 3ware controller on Linux 
2.6.15.4. Everything is in the Thomas-Krenn SR2400 server running on Debian.

3WARE DEVICE:
driver: 1.26.02.001
model: 8006-2LP
firmware: FE8S 1.05.00.068
bios: BE7X 1.08.00.048

DISKS:
2x Seagate ST3300831AS SATA 300GB

I used tw_cli (AMCC/3ware CLI (version 2.00.03.013)) to create RAID1 unit. 
After running cfdisk raid degraded and it can't be rebuilded. 
Next I tried to make JBOD units for testing. A 3ware driver was compiled into 
the kernel and also as a module. 

ERRORS:
Apr 26 15:07:10 krenn kernel: 3w-xxxx: scsi2: Command failed: status = 0xc7, 
flags = 0xc, unit #0.
Apr 26 15:07:14 krenn kernel: 3w-xxxx: scsi2: AEN: ERROR: Drive error: Port 
#0.
Apr 26 15:07:15 krenn kernel: 3w-xxxx: scsi2: Command failed: status = 0xc7, 
flags = 0xc, unit #0.
Apr 26 15:07:20 krenn kernel: 3w-xxxx: scsi2: Command failed: status = 0xc7, 
flags = 0xc, unit #0.
Apr 26 15:07:25 krenn kernel: 3w-xxxx: scsi2: Command failed: status = 0xc7, 
flags = 0xc, unit #0.
Apr 26 15:07:30 krenn kernel: 3w-xxxx: scsi2: Command failed: status = 0xc7, 
flags = 0xc, unit #0.
Apr 26 15:07:35 krenn kernel: 3w-xxxx: scsi2: Command failed: status = 0xc7, 
flags = 0xc, unit #0.
Apr 26 15:07:38 krenn kernel: sd 2:0:0:0: WARNING: Command (0x28) timed out, 
resetting card.

It looks like some incompatibility with driver. Cables and disks are new, no 
badblocks. Tommorow I will try to compile driver from 3ware website with 2.4 
kernel. 

Thanks in advance. 

-- 
 Marcin Hlybin, marcin.hlybin@swmind.com
 Sys/Net Administrator, tel. +48 12 2523 402

 SoftwareMind, www.softwaremind.pl | Where quality meets the future

Re: 3ware 8006-2LP on Linux 2.6 drive error, seagate disks

[Marcin Hlybin]

On Wednesday 26 of April 2006 17:32, Marcin Hlybin wrote:

> Apr 26 15:07:10 krenn kernel: 3w-xxxx: scsi2: Command failed: status = 0xc7, 
> flags = 0xc, unit #0.

Ok, I've solved my problem by changing SATA cables. I have tested controller 
with massive I/O operations with both enabled and disabled write-caching. 
Admittedly next day TK server crashed with NMI, parity errors and so on, but 
now I have new mainboard, new riser and everything works just fine. 
I use tw_cli to maintain the RAID. 

Regards

-- 
 Marcin Hlybin, marcin.hlybin (at) swmind.com
 Sys/Net Administrator, tel. +48 12 2523 402

 SoftwareMind, www.softwaremind.pl | Where quality meets the future