Ben's notes

These steps below verifies the consistency of RAID, but not if the contents of data is correct!

• Find the RAID (md) device you want to check:

  cat /proc/mdstat 
  Personalities : [linear] [raid0] [raid1] [raid10] [raid6] [raid5] [raid4] 
  md2 : active raid6 sda3[0] sdgd3[7] sdgc3[6] sdgb3[5] sdga3[4] sdd3[8] sdc3[2] sdb3[1]
        11692753536 blocks super 1.2 level 6, 64k chunk, algorithm 2 [8/8] [UUUUUUUU]
        
  md1 : active raid1 sda2[0] sdb2[1] sdc2[2] sdd2[3]
        2097088 blocks [4/4] [UUUU]
        
  md0 : active raid1 sda1[0] sdb1[1] sdc1[2] sdd1[3]
        2490176 blocks [4/4] [UUUU]
        
  unused devices: <none>

• Start scrub process:

  echo check > /sys/block/md2/md/sync_action

• This will scrub your array and count any bit mismatches in:

  cat /sys/block/md2/md/mismatch_cnt

• Then you can repair the mismatching blocks with:

  echo repair > /sys/block/md2/md/sync_action

• Monitor progress:

  cat /proc/mdstat
  Personalities : [linear] [raid0] [raid1] [raid10] [raid6] [raid5] [raid4] 
  md2 : active raid6 sda3[0] sdgd3[7] sdgc3[6] sdgb3[5] sdga3[4] sdd3[8] sdc3[2] sdb3[1]
        11692753536 blocks super 1.2 level 6, 64k chunk, algorithm 2 [8/8] [UUUUUUUU]
        [>....................]  check =  0.0% (109008/1948792256) finish=893.6min speed=36336K/sec
        
  md1 : active raid1 sda2[0] sdb2[1] sdc2[2] sdd2[3]
        2097088 blocks [4/4] [UUUU]
        
  md0 : active raid1 sda1[0] sdb1[1] sdc1[2] sdd1[3]
        2490176 blocks [4/4] [UUUU]
        
  unused devices: <none>