Both drives already had PM disabled, visible in hdparm -i: "AdvancedPM=yes: disabled (255) WriteCache=enabled" Looking at the smart reporting, it is showing both drives have a FAILING_NOW condition for Seek_Error_Rate. I don't know what to believe, because it seems like whatever drives I attach to this system are chewed up and start showing Seek_Error_Rate failure conditions. /dev/sda: 7 Seek_Error_Rate 0x000b 046 046 067 Pre-fail Always FAILING_NOW 393853 /dev/sdb: 7 Seek_Error_Rate 0x000b 044 044 067 Pre-fail Always FAILING_NOW 2556544 I swapped in 2 more drives of the same model, and one exhibits the same Seek_Error_Rate FAILING_NOW condition. I now have 4 out of 5 of this same model drive which are failing. They appear to be from the same batch, so I'm not ruling out some kind of manufacturing defect, but this definitely seems strange. I guess I'm just fishing to see if there is anything on the system that could have damaged the drives. Thanks, -Andrew On Jan 27, 2008 1:33 PM, Jim Paris <jim@xxxxxxxx> wrote: > Andrew Paprocki wrote: > > I've been noticing something strange on an AMD Geode LX board that I > > have.. I have two SATA drives connected to the onboard Sil3114 chip, > > and the drives appear to be continually restarting (soft resetting?) > > during normal operation when nothing at all is happening on the > > machine. You can hear the drives doing it as well as feel it > > physically if you touch the drive. They are spinning down and back up > > again over and over again. All the while the OS never prints out any > > ata/scsi problems. The only manifestation of this in the kernel is > > that if you're doing something w/ the drives, it pauses momentarily > > while this happens (for instance, during an ext3 format). > > It could be drive power management. Try "hdparm -B 255" or "hdparm -B > 254" to turn that off. The output of "smartctl -A" output can also be > helpful to figure out what's causing it. > > -jim > > - To unsubscribe from this list: send the line "unsubscribe linux-ide" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
