Re: SATA eating my disk, port reset, destroying unrelated data

[Robert Hancock <hancockr@shaw.ca>]

Norbert Preining wrote:
> Dear all!
>
> (please Cc me for answers)
>
> Since about 5 days I am having serious problems with my SATA drive:
>
> kernel 2.6.22 (from Debian/sid)
> hardware nv
>
> Sometimes at boot time, often/always at disk io intense stuff:
>
> ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x400000 action 0x2

Serror 0x400000 means a handshake error. Usually Serror indications are 
due to a hardware problem (bad SATA cable, power or drive problem).

> ata1.00: (BMDMA stat 0x25)
> ata1.00: cmd 35/00:00:2a:6f:c0/00:04:0c:00:00/e0 tag 0 cdb 0x0 data 524288 out
>          res 51/84:10:1a:72:c0/84:01:0c:00:00/e0 Emask 0x10 (ATA bus error)
> ata1: soft resetting port
> ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
> ata1.00: configured for UDMA/133
> ata1: EH complete
>
> Even worse, sometimes the reset does not work ...
>
> ata1: device not ready (errno=-16), forcing hardreset
> ata1: hard resetting port
> ata1 SRST failed (errno=-19)
> ata1: reset failed (errno=-19), retrying in 10 secs
> ..
>
> (typed from a digital photo, nothing remains in the logs)
>
> After this I need to do a cold boot otherwise the drive is really in a
> bad state and not even the bios gets it right.

If even the BIOS cannot reset properly then that also really points to a 
hardware problem..

> Interestingly the whole stuff DID work for a long time until I did too
> many things at the same time: 2 x svn up, copying 40G from the SATA
> drive to an USB drive, aptitude upgrade. Before I did regularly the same
> stuff (like svn up etc), but this time it was too much, it seems.
>
> Apropos data hosing: After the first incident some data on my windows
> partitions (/dev/sda1) was hosed, programs missing, chkdisk necessary
> etc.
>
> I attach dmesg (from the current boot with a succeeding soft reset, I
> interrupted the svn process before the SATA drives goes into hard reset
> failures), .config, lspci -v output.
>
> Are there any chances that using 2.6.23 will improve/fix this? Any other
> suggestions?
>
> I would consider it an hardware problem, but since it started at one big
> io thingy and is persistent since then I am a bit sceptic.

--
Robert Hancock      Saskatoon, SK, Canada
To email, remove "nospam" from hancockr@nospamshaw.ca
Home Page: http://www.roberthancock.com/

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/