ext4 with LVM and RAID5 (3 disks)

Single LVM on top of a 3 disk RAID5 array, formatted as ext4

Download VMDK (zipped)

https://files.iblue.team/279b6e00-851e/2GB-ext4-raid5/LINUX-MINT-8.vmdk.7z https://files.iblue.team/279b6e00-851e/2GB-ext4-raid5/LINUX-MINT-7.vmdk.7z https://files.iblue.team/279b6e00-851e/2GB-ext4-raid5/LINUX-MINT-6.vmdk.7z

Download RAW (zipped)

https://files.iblue.team/279b6e00-851e/2GB-ext4-raid5/LINUX-MINT-8.vmdk.dd.7z https://files.iblue.team/279b6e00-851e/2GB-ext4-raid5/LINUX-MINT-7.vmdk.dd.7z https://files.iblue.team/279b6e00-851e/2GB-ext4-raid5/LINUX-MINT-6.vmdk.dd.7z

I've created a 3 disk RAID5 array using mdadm/LVM, which contained a few files to demonstrate how data is striped across an array.

How it was created;

1. Create file of a fixed size, using a unique string/word which is easily identifiable such as 'APPLE'

The 'yes' command will output the string passed to it (APPLE) continually until killed. It's piped into head, and it'll output a file of a size of 1GB which is redirected to apple.txt

$ yes APPLE | head -c 1073741824 > apple.txt

Do this 3 more times using banana, carrot, date, eggplant, to produce 5 x 1GB text files.

1. Create 3 x 2GB volumes, attach to virtual machine.

2. Unwilling to restart the virtual machine, I probed the SCSI host to detect new disks.

$ for host in /sys/class/scsi_host/*; do echo "- - -" | sudo tee $host/scan; ls /dev/sd* ; done

$ lsblk
NAME   MAJ:MIN RM  SIZE RO TYPE MOUNTPOINTS
sda      8:0    0    2G  0 disk
sdb      8:16   0   80G  0 disk
├─sdb1   8:17   0    1M  0 part
├─sdb2   8:18   0  513M  0 part /boot/efi
└─sdb3   8:19   0 79.5G  0 part /
sdc      8:32   0    2G  0 disk
sdd      8:48   0    2G  0 disk

So we have sda, sdc, and sdd.

Create RAID5 volume using mdadm

$ mdadm --create --verbose /dev/md0 --level=5 --raid-devices=3 /dev/sda /dev/sdc /dev/sdd
mdadm: layout defaults to left-symmetric
mdadm: layout defaults to left-symmetric
mdadm: chunk size defaults to 512K
mdadm: size set to 2094080K
mdadm: Defaulting to version 1.2 metadata
mdadm: array /dev/md0 started.

Check block device output prior to creating filesystem

$ lsblk
NAME   MAJ:MIN RM  SIZE RO TYPE  MOUNTPOINTS
sda      8:0    0    2G  0 disk
└─md0    9:0    0    4G  0 raid5
sdb      8:16   0   80G  0 disk
├─sdb1   8:17   0    1M  0 part
├─sdb2   8:18   0  513M  0 part  /boot/efi
└─sdb3   8:19   0 79.5G  0 part  /
sdc      8:32   0    2G  0 disk
└─md0    9:0    0    4G  0 raid5
sdd      8:48   0    2G  0 disk
└─md0    9:0    0    4G  0 raid5
sr0     11:0    1 1024M  0 rom

$ fdisk -l /dev/md0
Disk /dev/md0: 3.99 GiB, 4288675840 bytes, 8376320 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 524288 bytes / 1048576 bytes

Create ext4 filesystem on /dev/md0

$ mkfs.ext4 /dev/md0
mke2fs 1.46.5 (30-Dec-2021)
Creating filesystem with 1047040 4k blocks and 262144 inodes
Filesystem UUID: 63a60555-ea65-4ed4-83f7-e7e851db8f52
Superblock backups stored on blocks:
        32768, 98304, 163840, 229376, 294912, 819200, 884736

Allocating group tables: done
Writing inode tables: done
Creating journal (16384 blocks): done
Writing superblocks and filesystem accounting information: done

Mount volume to /mnt/iblue5 and copy dummy data to it (attempted to copy all 5 x 1GB files, which partially failed due to insufficient space).

$ mkdir /mnt/iblue5
$ mount /dev/mapper/data-iblue /mnt/iblue5
$ ls /mnt/iblue5
lost+found
$ cp /home/user/Desktop/*.txt /mnt/iblue5/
$ ls -lah /mnt/iblue5
total 3.9G
drwxr-xr-x  3 root root 4.0K Dec  9 22:39 .
drwxr-xr-x 17 root root 4.0K Dec  9 22:25 ..
-rw-r--r--  1 root root 1.0G Dec  9 22:38 apple.txt
-rw-r--r--  1 root root 1.0G Dec  9 22:38 banana.txt
-rw-r--r--  1 root root 1.0G Dec  9 22:39 carrot.txt
-rw-r--r--  1 root root 858M Dec  9 22:39 date.txt
-rw-r--r--  1 root root    0 Dec  9 22:39 eggplant.txt
drwx------  2 root root  16K Dec  9 22:37 lost+found

Unmount /dev/md0

$ umount /dev/md0
$ ls /mnt/iblue5