[clug] SATA Drive errors in the log

[Eyal Lebedinsky]

Tomasz Ciolek wrote:
> Hi Guys
>
> Can anyone enlighten me if this is a drive thats dying or the sata controller thats playing up?
>
> Mar  4 19:00:21 citadel kernel: ata1: status=0x51 { DriveReady SeekComplete Error }
> Mar  4 19:00:21 citadel kernel: ata1: error=0x84 { DriveStatusError BadCRC }
> Mar  4 19:00:25 citadel kernel: ata1: no sense translation for status: 0x51
> Mar  4 19:00:25 citadel kernel: ata1: translated ATA stat/err 0x51/00 to SCSI SK/ASC/ASCQ 0x3/11/04
> Mar  4 19:00:25 citadel kernel: ata1: status=0x51 { DriveReady SeekComplete Error }
> Mar  4 19:00:27 citadel kernel: ata1: no sense translation for status: 0x51
> Mar  4 19:00:27 citadel kernel: ata1: translated ATA stat/err 0x51/00 to SCSI SK/ASC/ASCQ 0x3/11/04
>
> Tomasz

SATA is still less stable than PATA on Linux, some controllers are not really
production quality (see the amount of develolment even in the latest kernel).

To check the disk for real you should check the smart log. You can also run the
manufacturer's disk (non-destructive) health check utility (most offer it on
their web).

I saw some reports that suggest a high load on DMA may lead to failures on some
chipsets. Not too many reports though.

Finally, it seems that some controllers have high sensitivity to power quality
and reading the Linux-ide/Linux-raid lists I see many similar problems go away
once a better PS is installed.

If the smart and health-check are good, and you can provoke the problem at will
(say by running some high io setup) then try to disconnect some high drain
devices (e.g. a modern high-end video card, other disks etc.) and if the problem
goes away then I will suspect the PS.

Good luck.

-- 
Eyal Lebedinsky (eyal at eyal.emu.id.au) <http://samba.org/eyal/>
	attach .zip as .dat