Disk Partitioning and Filesystem Creation

When you buy a new physical hard drive (also called a disk or block device), it is essentially a large, flat, empty space. Your operating system cannot just start writing files to it willy-nilly. First, you need to create partitions. A partition is a logical section of the disk. Think of it as cutting a large pizza into slices - you are dividing the total available space into separate, independent pieces. Each partition can be used for a different purpose. For example, you might have one partition for the Linux operating system itself, another for user files, and a third for swap space (a special area used as virtual memory). This separation provides several benefits: it isolates problems (a full partition for user files does not crash the system partition), it allows different filesystems on different partitions, and it enables dual-booting multiple operating systems.

To create partitions, you use a tool like fdisk or parted. In the exam, you will use commands such as fdisk /dev/sda to start partitioning the first disk. Inside fdisk, you enter an interactive menu where you can create a new partition table (which is like the table of contents for the disk), then add, delete, and resize partitions. The partition table is stored in the Master Boot Record (MBR) or GUID Partition Table (GPT). MBR is the older standard, limited to four primary partitions and a maximum disk size of 2TB. GPT is the modern standard, supporting many more partitions (up to 128 on Linux) and disks larger than 2TB. The LFCS exam assumes you know the differences and that GPT is preferred for modern systems.

Once you have created partitions, they are still unusable for data storage. You must format each partition with a filesystem. A filesystem is the method and data structure that the operating system uses to control how data is stored and retrieved on a partition. It is like deciding the rules for organising files in a library: where the card catalogue goes, how books are shelved, and what metadata (author, title, date) is recorded. Common Linux filesystems include: - ext4: The default for many Linux distributions, reliable and feature-rich. - XFS: Excellent for large files and high performance. - Btrfs: Modern, with advanced features like snapshots and compression, but more complex. - swap: Not a filesystem for storing files, but a special format used for virtual memory.

To format a partition, you use the mkfs command. For example, mkfs.ext4 /dev/sda1 creates an ext4 filesystem on the first partition of the first disk. This command writes the filesystem structure (superblock, inode tables, block groups) to the partition, preparing it to hold files. After formatting, the partition still needs to be mounted to be accessible. Mounting attaches the partition to a directory in the existing Linux filesystem tree. You mount it using mount /dev/sda1 /mnt, which makes the contents of /dev/sda1 appear under the /mnt directory. To make the mount persistent across reboots, you add an entry to the /etc/fstab file, which tells the system which partitions to mount automatically at boot time.

Why does this all matter? Without partitioning and filesystem creation, a disk is a blank slate. The process is like building a house: first, you divide the land into rooms (partitions), then you install the electrical wiring and plumbing (filesystem), and finally, you turn on the lights (mount). Linux gives you total control over this process, and the LFCS exam tests your ability to execute these tasks correctly from the command line.

An IT professional at a small company is setting up a new Linux web server. The server has a single 500GB SSD. The administrator needs to partition this disk to separate the operating system, application data, and logs, ensuring that a runaway log file cannot fill the entire disk and crash the server.

The step-by-step process they follow is:

- Partition 1: 20GB, for the root filesystem (/) - Partition 2: 100GB, for application data (/var/www) - Partition 3: 10GB, for log files (/var/log) - Partition 4: remaining space (approx 370GB) for user home directories (/home) - They press 'w' to write the partition table and exit fdisk. - Next, they format each partition with the appropriate filesystem: - sudo mkfs.ext4 /dev/sda1 (root) - sudo mkfs.ext4 /dev/sda2 (app data) - sudo mkfs.ext4 /dev/sda3 (logs) - sudo mkfs.ext4 /dev/sda4 (home) - They create mount points (directories) for each partition: sudo mkdir /mnt/root /mnt/www /mnt/log /mnt/home. - They mount each partition temporarily: sudo mount /dev/sda1 /mnt/root, etc. - They copy the operating system files to /mnt/root, then use sudo blkid to get the UUID (Universally Unique Identifier) of each partition. They edit /mnt/root/etc/fstab to add permanent mount entries using the UUIDs, ensuring that the partitions mount correctly after every boot.

This setup means that if the web application generates excessive logs, only the /var/log partition fills up, and the server remains operational. The administrator also created a swap partition (optional) for virtual memory, which is formatted with mkswap and enabled with swapon. In a real business context, this kind of structured partitioning improves reliability, security (by isolating different types of data), and simplified backup strategies (you can back up only the /var/www partition).