Re: ata3.00: exception Emask 0x10 SAct 0x0 SErr 0x400100 action 0x6 frozen
On Sat, 09 Jun 2012 15:56:58 -0700, David Christensen wrote:
> I have a new machine with debian-6.0.5-amd64 and:
>
> Intel i7-2600S processor
> Intel DQ67SW motherboard
> Intel 520 Series 60 GB SSD
>
>
> I noticed error messages similar to the following example on the
> console:
>
> [ 131.742744] ata3.00: exception Emask 0x10 SAct 0x0 SErr 0x400100 action 0x6 frozen
> [ 131.742787] ata3.00: irq_stat 0x08000000, interface fatal error
> [ 131.742806] ata3: SError: { UnrecovData Handshk }
> [ 131.742822] ata3.00: failed command: WRITE DMA EXT
> [ 131.742839] ata3.00: cmd 35/00:00:c8:ce:79/00:04:00:00:00/e0 tag 0 dma 524288 out
> [ 131.742840] res 50/00:00:c8:ca:79/00:00:04:00:00/e0 Emask 0x10 (ATA bus error)
> [ 131.742886] ata3.00: status: { DRDY }
Mmm... given that the system is new (is a DIY or you bought the system
already assembled)? it can be a bad or poor sata cable or loose
connection. I'd first try to replace it and see how it goes.
> I assume these are error messages coming from the disk device driver
> (?). What do they mean?
It can be something coming from the kernel or a hardware problem. I'd
first locate the offending device and replace the SATA cable (always
use good quality cables and discard the ones coming with the motherboard).
dmesg | grep "ata3.00" will tell what's attached to the SATA 3 port.
> I then upgraded the kernel to 3.2.0-0.bpo.2-amd64 via debian-backports,
> and the errors filled the console whenever the disk was busy (such as
> running tasksel and installing Graphical desktop environment).
>
>
> STFW, I installed sysfsutils and added the following entries to
> /etc/sysfs.conf. The quantity of messages decreased:
>
> block/sda/device/queue_depth=1
> block/sdb/device/queue_depth=1
>
>
> Any suggestions?
Kernels hackers can tell you some tips (parameters) you can try to pass to
the kernel line, let's hope they are landing around here :-)
Greetings,
--
Camaleón
Reply to: