How to Resize an AWS EBS Volume Without Downtime

Why Resize an AWS EBS Volume Without Downtime?

When you first provision a cloud server, you make your best guess at how much storage you will need. But production environments are unpredictable. Logs accumulate. Databases grow. A sudden spike in traffic can push disk utilisation from a comfortable 40% to a worrying 85% in a matter of days.

The good news: modern cloud providers like AWS let you expand an EBS (Elastic Block Store) volume while the instance is fully running. No scheduled maintenance window. No SSH session drop. No service interruption.

This post walks through exactly how to do it — and more importantly, explains why each step is necessary.

When Should You Resize?

Here are common real-world triggers that signal it is time to expand your volume:

• Disk utilisation crosses 75–80% — a good early-warning threshold
• Your monitoring or observability stack starts throwing low disk space alerts
• A database optimisation or vacuum job is about to run and needs temporary headroom
• You are enabling a new service (logging agent, APM tool, backup daemon) that will write to disk
• A large deployment or migration is planned and you want a safety buffer
• Automated disk-usage reports show a steady upward trend — growth is inevitable

💡 Pro tip: Do not wait until you are at 90%+. Resize proactively when you hit 75–80%. Emergency resizes under pressure lead to mistakes.

Understanding the Three Layers

Before running any commands, it helps to understand that a disk has three distinct layers — and you need to expand all three in sequence:

Layer	What It Is	How You Expand It
1. The Disk	The raw EBS volume — just a block of storage AWS allocates	AWS Console / CLI
2. The Partition	A defined slice of the disk with a start and end sector	growpart
3. The Filesystem	The structure (ext4/xfs) that actually stores your files	resize2fs or xfs_growfs

Think of it like this: expanding the EBS volume is like buying a bigger plot of land. growpart moves the fence to claim the new land. resize2fs builds on it so you can actually use it.

New to these Linux tools? Our Linux Saviours guide covers the command-line essentials every server admin needs.

Step 1 — Expand the EBS Volume in AWS

Log into the AWS Console and navigate to EC2 → Elastic Block Store → Volumes. Select the volume attached to your instance, click Actions → Modify Volume, and enter the new size. Click Modify to confirm.

The change is applied within seconds to a few minutes. The instance keeps running throughout. You can verify the new size is visible at the OS level by running:

lsblk

A sample output after the AWS-side resize might look like this:

NAME         MAJ:MIN RM  SIZE RO TYPE MOUNTPOINTS
nvme0n1      259:0    0  300G  0 disk
├─nvme0n1p1  259:1    0   99G  0 part /
├─nvme0n1p15 259:3    0  106M  0 part /boot/efi
└─nvme0n1p16 259:4    0  913M  0 part /boot

Notice: the disk (nvme0n1) shows the new size, but the partition (nvme0n1p1) still shows the old size. This is expected — you have not run growpart yet.

Step 2 — Grow the Partition

Now tell the OS to extend the partition boundary to fill the new disk space. The command takes the disk name and partition number as separate arguments:

sudo growpart /dev/nvme0n1 1

If successful, you will see output like CHANGED: partition=1 start=... old: size=... end=... new: size=... end=... confirming the partition boundary was moved.

Step 3 — Extend the Filesystem

First, check your filesystem type:

df -T /

Then run the appropriate command based on the result:

growpart resize2fs — Expand ext4 on AWS EBS (Ubuntu, Debian)

sudo resize2fs /dev/nvme0n1p1

xfs_growfs — Resize AWS EBS Volume on XFS (Amazon Linux, RHEL)

sudo xfs_growfs -d /

Step 4 — Verify

Run the final check to confirm the OS now sees and can use the full expanded space:

df -h /

Expected output (example):

Filesystem      Size  Used Avail Use% Mounted on
/dev/root        295G   47G  248G  16% /

If the Size column reflects your new target size, you are done. No reboot required.

Important Constraints to Know

• You can only increase EBS volume size — shrinking is not supported
• After a modification, AWS enforces a 6-hour cooldown before you can modify the same volume again
• The OS-level steps (growpart + resize2fs) must be run manually — AWS does not do this automatically
• Always verify with lsblk and df -h before and after to confirm each layer expanded correctly

Summary

Resizing an EBS volume live is straightforward once you understand the three-layer model. AWS handles the disk. You handle the partition and filesystem. The whole process takes under five minutes and your application never knows it happened.

Do it proactively, do it calmly, and your servers will never run out of space at the worst possible moment. For more on keeping Linux servers healthy, see how we audited 41 Ubuntu servers for a critical CVE — the same discipline applies here.