History: RAID Maintenance Procedures

Source of version: 9 (current)

!!!Repairing Broken Sectors #Mark the device as failed in the array +{CODE()}mdadm <raid dev> -f <problem device>{CODE} #Remove the failed device from the array +{CODE()}mdadm <raid dev> -r <problem device>{CODE} #Make the drive read all its sectors to detect all the problems +{CODE()}dd if=<problem device> of=/dev/null conv=noerror{CODE} #With the problems known to the drive, erase all its data so it can do sector re-mapping +{CODE()}dd if=/dev/zero of=<problem device> bs=8192{CODE} #Verify the drive has no bad sectors now +{CODE()}smartctl -t long <problem device>{CODE} #Wait for test completion +{CODE()}smartctl -l selftest <problem device>{CODE} #Re-create the partition table +{CODE()}parted <known good RAID member> unit s print # Learn what a correct geometry looks like parted <problem device> mklabel gpt mkpart primary <start>s <end>s # Configure the correct geometry{CODE} #Make sure the system is aware of the newly created partition +{CODE()}partprobe <problem device>{CODE} #Add the device into the array +{CODE()}mdadm <raid dev> -a <problem device>{CODE} #Monitor RAID rebuild progress +{CODE()}cat /proc/mdstat{CODE} !!!Forcing Drive Detection #You have to know which SCSI host (bus) to scan, but once that is known: +{CODE()}echo "- - -" > /sys/class/scsi_host/host#/scan{CODE}

---