Re: promise sataII 150 TX4 issue

[Robert Hancock <hancockrwd@gmail.com>]

On 08/17/2010 05:39 PM, kenneth johansson wrote:
> 00:0d.0 Mass storage controller: Promise Technology, Inc. PDC20518/
> PDC40518 (SATAII 150 TX4) (rev 02)
>
> I have a raid5 with 3 disks in and have started so see some weired
> problems.First I thought it was the disk since it was always the same but
> after changing the disk I now have the same problems but another disk so
> I'm starting to suspect a driver issue.
>
> The only disk I have in the system is connected to this pci card and the
> problem is always with the disk named sda regardless if it is connected
> to the first sata port or the second and first is empty.
> I have switch power cables and sata cables around but problem is not
> moving from sda.
>
> problem seen on kernel version vmlinuz-2.6.35,vmlinuz-2.6.33.3. was
> running vmlinuz-2.6.28.7 for about a year before that without any problem.
>
> Now it took about two month before sda got kicked out of the raid on
> 2.6.33.3 but 2.6.35 have not lasted more than a few days.
>
> -------------------------------------------
> Aug 15 20:27:22 amd kernel: [171585.020040] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x180000 action 0x6 frozen
> Aug 15 20:29:15 amd kernel: [171585.020080] ata1: SError: { 10B8B Dispar }
> Aug 15 20:29:15 amd kernel: [171585.020100] ata1.00: failed command: WRITE DMA EXT
> Aug 15 20:29:15 amd kernel: [171585.020125] ata1.00: cmd 35/00:f8:3f:bb:a1/00:03:73:00:00/e0 tag 0 dma 520192 out
> Aug 15 20:29:15 amd kernel: [171585.020127]          res 40/00:01:00:4f:c2/00:00:00:00:00/00 Emask 0x4 (timeout)

The controller's reporting errors on the SATA link. I would still tend 
to suspect some kind of hardware issue - maybe that port connector is 
bad on the card or something? I would be surprised if a driver fault 
could cause this to be reported spuriously.