Re: Did my SATA drive fail? What should I do?

[Robert Hancock]

Nathan wrote:
> First off:  Hi!  I'm new to this list.  Googling some of the errors in
> my logs brought up several posts to this list, none of which led to
> any obvious resolutions, so I thought I'd try joining the list...
>
> I have a linux server[1] that has been acting as an asterisk pbx for
> about three months without a problem.  Today, the SATA hard drive
> suddenly went into read-only mode (!?).  Reading seems to work just
> fine, as I've ssh'd in and looked at a bunch of files, including
> /var/log/messages [2].
>
> Is this just my hardware failing?  Could this a kernel bug?
> Recommendations?  I would just reboot the thing, but asterisk is still
> successfully handling dozens of calls right now, despite not being
> able to write to the disk...
>
> [1] # uname -a
> Linux phoneserver3 2.6.25-gentoo-r7 #2 SMP Thu Oct 16 20:59:53 MDT
> 2008 x86_64 Intel(R) Xeon(R) CPU E5410 @ 2.33GHz GenuineIntel
> GNU/Linux
>
> [2] (last several lines from /var/log/messages)
>
> Jan 22 14:43:09 phoneserver3 dhcpd: DHCPDISCOVER from
> 00:04:f2:1e:7e:21 via eth2: network 10.254.254/24: no free leases
> Jan 22 14:43:09 phoneserver3 dhcpd: DHCPDISCOVER from
> 00:04:f2:1e:7c:6b via eth2: network 10.254.254/24: no free leases
> Jan 22 14:43:35 phoneserver3 dhcpd: DHCPDISCOVER from
> 00:04:f2:1e:0d:47 via eth2: network 10.254.254/24: no free leases
> Jan 22 14:43:36 phoneserver3 dhcpd: DHCPDISCOVER from
> 00:1f:28:83:00:80 via eth2: network 10.254.254/24: no free leases
> Jan 22 14:43:41 phoneserver3 ata1.00: exception Emask 0x0 SAct 0x0
> SErr 0x0 action 0x0
> Jan 22 14:43:41 phoneserver3 ata1.00: BMDMA stat 0x25
> Jan 22 14:43:41 phoneserver3 ata1.00: cmd
> ca/00:20:00:7c:ef/00:00:00:00:00/ef tag 0 dma 16384 out
> Jan 22 14:43:41 phoneserver3 res 51/10:20:00:7c:ef/00:00:00:00:00/ef
> Emask 0x81 (invalid argument)
> Jan 22 14:43:41 phoneserver3 ata1.00: status: { DRDY ERR }
> Jan 22 14:43:41 phoneserver3 ata1.00: error: { IDNF }

Most likely this is the disk failing. It's reporting the sector wasn't 
found during a write operation.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/