Expanding Storage on RHEL Using Logical Volume Management (LVM)

The Problem: Fixed Partition Limitations

Standard disk partitions are rigid. If a database or application log directory fills up, expanding a standard partition often requires unmounting the drive, risking downtime, or moving data to a new disk entirely. Red Hat Enterprise Linux (RHEL) uses Logical Volume Management (LVM) to solve this by adding a layer of abstraction between the physical hardware and the filesystem.

The primary takeaway is that LVM allows you to pool multiple physical disks into a single storage reservoir, enabling you to resize volumes online without taking services offline.

LVM Architecture Overview

To manage LVM, you must understand the three-layer hierarchy:

• Physical Volume (PV): A raw disk or partition (e.g., /dev/sdb) initialized for LVM use.
• Volume Group (VG): A pool created by combining one or more PVs. Think of this as a virtual disk.
• Logical Volume (LV): A slice of the VG that acts as a virtual partition where you format a filesystem (e.g., XFS or EXT4).

Prerequisites

• A RHEL system (Version 7, 8, or 9) with root or sudo privileges.
• An unallocated physical disk or a new virtual disk attached to the system.
• Knowledge of the target mount point that requires more space.

Procedure: Expanding a Logical Volume

This guide covers the scenario where you have added a new physical disk to a server and need to extend an existing volume.

Step 1: Initialize the New Physical Volume

Run this command on the host terminal to prepare the new disk for LVM. Replace /dev/sdb with your actual device identifier.

sudo pvcreate /dev/sdb

Risk: Running this on a disk containing data will overwrite the partition table.

Step 2: Extend the Volume Group

Add the new PV to your existing Volume Group. Use vgs to find the name of your VG (e.g., vg_data).

sudo vgextend vg_data /dev/sdb

Step 3: Expand the Logical Volume

Increase the size of the specific LV. The following example adds 50GB to the volume named lv_storage.

sudo lvextend -L +50G /dev/vg_data/lv_storage

Alternatively, to use all remaining free space in the VG, use -l +100%FREE instead of -L +50G.

Step 4: Resize the Filesystem

The LV is now larger, but the filesystem inside it does not yet recognize the new space. RHEL defaults to XFS, which supports online growing.

sudo xfs_growfs /mnt/data

Note: If you are using EXT4, use sudo resize2fs /dev/vg_data/lv_storage.

Comparison: XFS vs. EXT4 in LVM

Feature	XFS (RHEL Default)	EXT4
Online Expansion	Supported	Supported
Online Shrinking	Not Supported	Supported (Offline)
Growth Command	xfs_growfs	resize2fs

Verification and Diagnostics

To confirm the operation was successful, run these checks in order:

1. Physical Layer: Run sudo pvs to ensure the new disk is listed and assigned to the correct VG.
2. Group Layer: Run sudo vgs to verify the total pool size has increased.
3. Logical Layer: Run sudo lvs to see the new size of the specific LV.
4. Filesystem Layer: Run df -h /mnt/data. The "Size" column should reflect the added capacity.

Limitations and Recovery

Snapshot Space: Do not allocate 100% of your Volume Group to Logical Volumes if you plan to use LVM snapshots. Snapshots require unallocated space in the VG to store the "diff" data between the original volume and the snapshot.

Rollback:

LVM expansion is a non-destructive additive process. However, if you added a disk by mistake, you can remove it only if the disk contains no active data chunks. Use pvmove to migrate data off the disk first, then vgreduce to remove the PV from the group.