 koitsuPremium,MVM
join:2002-07-16
Mountain View, CA
kudos:19
1 edit | said by rustydusty:Replaced all drives, works for a bit then shows critical. Pulled the "bad" drive and put it in a different machine and works great. I'd like to see SMART attributes for all of the disks in this RAID-10 array, preferably from smartmontools 6.0 (smartctl -x output). There may be something disk-level that's triggering the array to mark the drive as bad (such as an I/O request taking too long). I'm dealing with a user in this exact situation, though on an nVidia controller, in another thread.
smartmontools does not have support for AMD's driver shims (i.e. has no idea if ATA passthrough is offered or not) so you would need to put the disks on a different system attached to SATA ports that are non-RAID. Alternately I guess you could try switching from RAID mode to AHCI mode in your BIOS (so that they appear as individual/standard disks to the underlying driver subsystem, thus smartctl -x /dev/sda (for the first disk), smartctl -x /dev/sdb (for the 2nd disk), etc. would work). I would strongly suggest doing a full backup before trying this.
You can use smartctl --scan to get a list of all the /dev strings and arguments to pass smartctl. Also, if on Windows Vista or newer, make sure you try all of this running Command Prompt as Administrator.
--
Making life hard for others since 1977.
I speak for myself and not my employer/affiliates of my employer. |
