These instructions assume that the steps to assemble the hardware components of the RAID has been completed.

The steps are to be performed via ssh.

1. Identify the block (two 3.7TB drives in this example):

   root@turris:~# lsblk
   
   NAME         MAJ:MIN RM  SIZE RO TYPE  MOUNTPOINT
   sda            8:0    0  3.7T  0 disk              <-- device 1
   sdb            8:16   0  3.7T  0 disk              <-- device 2
   mtdblock0     31:0    0    1M  0 disk  
   mtdblock1     31:1    0    7M  0 disk  
   mmcblk0      179:0    0  7.3G  0 disk  
   `-mmcblk0p1  179:1    0  7.3G  0 part  /
   mmcblk0boot0 179:8    0    4M  1 disk  
   mmcblk0boot1 179:16   0    4M  1 disk  
   mmcblk0rpmb  179:24   0    4M  0 disk  

2. Disable NCQ:

   root@turris:~# echo 1 > /sys/block/[device 1]/device/queue_depth
   root@turris:~# echo 1 > /sys/block/[device 2]/device/queue_depth 

3. Create array ([device 1] → sda; [device 2] → sdb):

   root@turris:~# mdadm --create --verbose /dev/md0 --level=mirror --raid-devices=2 /dev/sda /dev/sdb

4. Monitor status:

   root@turris:~# cat /proc/mdstat

5. Create file system (btrfs in this example):

   root@turris:~# mkfs.btrfs /dev/md0

6. Find UUID for /dev/md0:

   root@turris:~# lsblk -o NAME,SIZE,MOUNTPOINT,STATE,FSTYPE,MODEL,UUID
   
   NAME          SIZE MOUNTPOINT STATE   FSTYPE            MODEL            UUID
   sda           3.7T            running linux_raid_member ST4000DM005-2DP1 f4b01a9e-7349-057d-f5f2-61ccbd78b1ea
   `-md0         3.7T                    btrfs                              6282e17f-2797-49dc-b4f9-e9452c35b922  <-- UUID
   sdb           3.7T            running linux_raid_member ST4000DM005-2DP1 f4b01a9e-7349-057d-f5f2-61ccbd78b1ea
   `-md0         3.7T                    btrfs                              6282e17f-2797-49dc-b4f9-e9452c35b922  <-- UUID
   mtdblock0       1M                                                       
   mtdblock1       7M                                                       
   mmcblk0       7.3G                                                       
   `-mmcblk0p1   7.3G /                  btrfs                              1ecd23ea-3018-4a83-9072-d337b0d00c5c
   mmcblk0boot0    4M                                                       
   mmcblk0boot1    4M                                                       
   mmcblk0rpmb     4M                                                       

7. Mount drive by adding a section (choose one of the two options) to /etc/config/fstab:

   # cat /etc/config/fstab
   
   # mount using UUID
   config mount
   	option enabled '1'
   	option uuid '[UUID]'
   	option target '/mnt/raid'
   
   # mount using device 
   config mount
   	option enabled '1'
   	option device '/dev/md0'
   	option target '/mnt/raid'
   

8. Create the settings file (/etc/mdadm/mdadm.conf) with the following contents:

   # mdadm.conf
   #
   # Please refer to mdadm.conf(5) for information about this file.
   #
   
   DEVICE partitions
   
   # auto-create devices with Debian standard permissions
   CREATE owner=root group=disk mode=0660 auto=yes
   
   # automatically tag new arrays as belonging to the local system
   HOMEHOST <system>
   
   # instruct the monitoring daemon where to send mail alerts
   MAILADDR root
   
   # definitions of existing MD arrays

9. Append the details of the RAID to the settings file:

   # mdadm --examine --scan >> /etc/mdadm/mdadm.conf

10. The file should resemble:

    # mdadm.conf
    #
    # Please refer to mdadm.conf(5) for information about this file.
    #
    
    DEVICE partitions
    
    # auto-create devices with Debian standard permissions
    CREATE owner=root group=disk mode=0660 auto=yes
    
    # automatically tag new arrays as belonging to the local system
    HOMEHOST <system>
    
    # instruct the monitoring daemon where to send mail alerts
    MAILADDR root
    
    # definitions of existing MD arrays
    ARRAY /dev/md/0 metadata=1.2 name=turris:0 UUID=69c415bb::########::########:########

11. Test the settings file:

    # mdadm --monitor --scan --test --oneshot

12. An email should be received:

    This is an automatically generated mail message from mdadm running on turris 
     
    A DegradedArray event had been detected on md device /dev/md/0. 
     
    Faithfully yours, etc. 
     
    P.S. The /proc/mdstat file currently contains the following: 
     
    Personalities : [linear] [raid0] [raid1] [raid10] [raid6] [raid5] [raid4] [multipath] 
    md0 : active raid1 sda[2] sdb[1] 
        1953383360 blocks super 1.2 [2/1] [_U] 
        [===>.................] recovery = 17.9% 
    (351595072/1953383360) finish=242.1min speed=110263K/sec 
     
    unused devices: <none> 

13. Adjust /etc/rc.local to enable RAID on boot:

    # disable NCQ
    echo 1 > /sys/block/sda/device/queue_depth
    echo 1 > /sys/block/sdb/device/queue_depth
    
    # start RAID array
    mdadm --assemble --scan
    
    # start RAID monitoring
    # --scan        -s   : find mail-address/program in config file (/etc/mdadm/mdadm.conf)
    # --test        -t   : Generate a TestMessage event against each array at startup
    # --daemonise   -f   : Fork and continue in child, parent exits
    mdadm --monitor --scan --test --daemonize
    
    # existing code
    exit 0

• Rebuild mdadm RAID array