[clug] ATA errors.

Peter Barker pbarker at barker.dropbear.id.au
Thu Aug 12 20:36:56 MDT 2010


On Fri, 13 Aug 2010, Andrew Janke wrote:

> I have a (ubuntu) machine that is occasionally freezing/hard locking,
> I have managed to track this via logging the syslog to another machine
> to this:
>
> Aug 13 10:17:28 wocket kernel: [ 5922.049325] ata5.00: exception Emask
> 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen
> Aug 13 10:17:28 wocket kernel: [ 5922.049332] ata5.00: cmd
> ca/00:08:7f:91:77/00:00:00:00:00/e1 tag 0 dma 4096 out
> Aug 13 10:17:28 wocket kernel: [ 5922.049332]          res
> 40/00:00:00:4f:c2/00:00:00:00:00/00 Emask 0x4 (timeout)
> Aug 13 10:17:28 wocket kernel: [ 5922.049334] ata5.00: status: { DRDY }
> Aug 13 10:17:28 wocket kernel: [ 5922.049340] ata5.00: hard resetting link
>
> So it would seem a dead/dying disk/cable/controller.
>
> The question is how do I map ata5 to a physical disk? (sd{a,b,c)?) are
> they just sequential?  Things like dmesg and lshw don't seem that
> informative.

I've wondered that myself on occasion.

https://ata.wiki.kernel.org/index.php/Libata_error_messages

Next question (from reading that page) - what is a "port number" according 
to libata :-)

If your devices are distinguishable on model number, there's enough 
information in the dmesg output I've looked at to determine the device.

You may also be able to examine the SMART information?

> Andrew Janke

Yours,
-- 
Peter Barker                          |   Programmer,Sysadmin,Geek.
pbarker at barker.dropbear.id.au	      |   You need a bigger hammer.
:: It's a hack! Expect underscores! - Nigel Williams


More information about the linux mailing list