On Tue, 4 Jul 2006 23:50:25 +0200
Manuel Bouyer <bouyer@antioche.eu.org> wrote:
> On Tue, Jul 04, 2006 at 02:59:48PM +0200, Lars Nordlund wrote:
> > Jul  4 13:48:07 glengoyne /netbsd: satalink0:0:0: wait timed out
> > Jul  4 13:48:07 glengoyne /netbsd: wd0h: device timeout reading fsbn 136892512 o f 136892512-136892639 (wd0 bn 199803052; cn 198217 tn 5 sn 1), retrying
> 
> This is not the usual mode of failure (it usually issue an ID not found),
> and AFAIK it's caught by the regex (ST3[2-9][0-9][0-9][0-9][0-9][0-9][A-Z]*).

Ahhhh.. I mixed it up with the 36GiB Seagate SCSI with the tag
ST336752LW. That was why it was one number short! Doh! Sorry about that.

> Are you sure this is not a WD_QUIRK_SPLIT_MOD15_WRITE issue ?

I did a fair bit of hammering on the drive afterwards and it seemed
stable. I had some use-cases which almost always triggered the problem.

Doing some testing with SPLIT_MOD15_WRITE now. I will let you know in a
couple of days.


Best regards,
	Lars Nordlund