0

I am running a server that connects to a multi-drive disk array, It is connected in a dual path configuration. Because of this, all of the drives show up twice.

I suffered a power outage and now the array is not assembling. I do however have the original commands I use. When I have attempted to assemble it says that there is no superblock. 

mdadm: No super block found on /dev/mapper/35000c500415cbb87 (Expected magic a92b4efc, got 00000000)
mdadm: no RAID superblock on /dev/mapper/35000c500415cbb87

I have two questions:

  1. Is it possible to reassemble the array at this point or is any data lost?
  2. How should I do this properly to avoid this in the future.

Initial commands run:

-Attempted to setup array, realized that I had twice as many drives. -Learned about multipath -Built a config that included the excluded the non-multipath drivers (see below) - Created a Raid6 array using the /dev/mapper/drives. Commands:

sudo mdadm --create --verbose /dev/md0 --level=6 --raid-devices=12 /dev/mapper/35000c500415cbb87 /dev/mapper/35000c500415cbbef /dev/mapper/35000c500415cfea3 /dev/mapper/35000c500415d0e3b /dev/mapper/35000c500415d6e3b /dev/mapper/35000c500415e456b /dev/mapper/35000c500415e4e5f /dev/mapper/35000c500415e8223 /dev/mapper/35000c500415ee867 /dev/mapper/35000c5004160b2df /dev/mapper/35000c5004160b6bb /dev/mapper/35000c50041611daf 

sudo mdadm -A /dev/mapper/35000c500415cbb87 /dev/mapper/35000c500415cbbef /dev/mapper/35000c500415cfea3 /dev/mapper/35000c500415d0e3b /dev/mapper/35000c500415d6e3b /dev/mapper/35000c500415e456b /dev/mapper/35000c500415e4e5f /dev/mapper/35000c500415e8223 /dev/mapper/35000c500415ee867 /dev/mapper/35000c5004160b2df /dev/mapper/35000c5004160b6bb /dev/mapper/35000c50041611daf

sudo mkfs.ext4 -F /dev/md0
sudo mkdir -p /mnt/md0
sudo mount /dev/md0 /mnt/md0
sudo mdadm --detail --scan | sudo tee -a /etc/mdadm/mdadm.conf
sudo update-initramfs -u

At that point, I setup folders and started using the drive, until the power outage shut it down. Upon restart, I cannot get it to reassemble.

/etc/multipath.conf

defaults {
        user_friendly_names no
}
blacklist {
    wwid "600508e000000000997d2e0655fb5108"
    wwid "5002538d408c9715"
    device {
        vendor "Dell    "
        product "Virtual Disk    "
    }
}

/etc/mdadm/mdadm.conf

# wildcards if desired.
#DEVICE partitions containers

DEVICE /dev/mapper/35000c500415cbb87
DEVICE /dev/mapper/35000c500415cbbef
DEVICE /dev/mapper/35000c500415cfea3
DEVICE /dev/mapper/35000c500415d0e3b
DEVICE /dev/mapper/35000c500415d6e3b
DEVICE /dev/mapper/35000c500415e456b
DEVICE /dev/mapper/35000c500415e4e5f
DEVICE /dev/mapper/35000c500415e8223
DEVICE /dev/mapper/35000c500415ee867
DEVICE /dev/mapper/35000c5004160b2df
DEVICE /dev/mapper/35000c5004160b6bb
DEVICE /dev/mapper/35000c50041611daf


# 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

# This configuration was auto-generated on Wed, 25 Jul 2018 23:01:52 +0000 by mkconf
ARRAY /dev/md0 metadata=1.2 name=media:0 UUID=5cfbe80c:d11c6b57:4beaff64:ed3bc6dd
Patch
  • 11
  • 4

1 Answers1

0

The problem was due to using the devices and not targeting partitions.

First step was to pre-partition each of the drives.

sudo mkfs.ext4 -F /dev/mapper/35000c500415cbb87
sudo mkfs.ext4 -F /dev/mapper/35000c500415cbbef
sudo mkfs.ext4 -F /dev/mapper/35000c500415cfea3
sudo mkfs.ext4 -F /dev/mapper/35000c500415d0e3b
sudo mkfs.ext4 -F /dev/mapper/35000c500415d6e3b
sudo mkfs.ext4 -F /dev/mapper/35000c500415e456b
sudo mkfs.ext4 -F /dev/mapper/35000c500415e4e5f
sudo mkfs.ext4 -F /dev/mapper/35000c500415e8223
sudo mkfs.ext4 -F /dev/mapper/35000c500415ee867
sudo mkfs.ext4 -F /dev/mapper/35000c5004160b2df
sudo mkfs.ext4 -F /dev/mapper/35000c5004160b6bb
sudo mkfs.ext4 -F /dev/mapper/35000c50041611daf

Then I was able to go through the same create process using the partitioned drives

sudo mdadm --create --verbose /dev/md0 --level=6 --raid-devices=12 /dev/dm-0 /dev/dm-1 /dev/dm-2 /dev/dm-3 /dev/dm-4 /dev/dm-5 /dev/dm-6 /dev/dm-7 /dev/dm-8 /dev/dm-9 /dev/dm-10 /dev/dm-11
sudo mkfs.ext4 -F /dev/md0
sudo mkdir -p /mnt/md0
sudo mount /dev/md0 /mnt/md0
sudo mdadm --detail --scan | sudo tee -a /etc/mdadm/mdadm.conf
sudo update-initramfs -u

Then I had to add to the /etc/mdadm/mdadm.conf instead of the /dev/mapper entries

DEVICE /dev/dm-0
DEVICE /dev/dm-1
DEVICE /dev/dm-2
DEVICE /dev/dm-3
DEVICE /dev/dm-4
DEVICE /dev/dm-5
DEVICE /dev/dm-6
DEVICE /dev/dm-7
DEVICE /dev/dm-8
DEVICE /dev/dm-9
DEVICE /dev/dm-10
DEVICE /dev/dm-11
Patch
  • 11
  • 4