AWS EC2 EBS Disk Expansion and Live Filesystem Resizing Without Downtime

When an AWS EC2 instance runs low on disk storage, AWS Elastic Block Store (EBS) Elastic Volumes allow administrators to increase volume capacity dynamically without detaching volumes or rebooting the server. However, expanding the volume in AWS only changes the underlying virtual disk size; you must still extend the partition table and resize the filesystem inside Linux.

Production Warning: Always create an AWS Backup snapshot or EBS snapshot before resizing partitions on mission-critical databases or production servers.

At a Glance

1. Modify the EBS volume size using the AWS Management Console or AWS CLI.
2. Monitor volume state until it transitions to in-use (optimizing).
3. SSH into the Linux instance and check disk block device visibility with lsblk.
4. Extend the partition boundary using growpart.
5. Resize the underlying filesystem using xfs_growfs (for XFS) or resize2fs (for ext4).
6. Verify the expanded capacity with df -h.

---

Prerequisites

Before resizing your storage, verify:

• IAM permissions to execute ec2:ModifyVolume and ec2:DescribeVolumes.
• SSH root or sudo access to the target EC2 Linux instance.
• Volume is formatted with an online-resizable filesystem (XFS or ext4).
• The EBS volume has not been modified within the last 6 hours (AWS Elastic Volume rate limit).

---

Step 1: Modify the EBS Volume in AWS

You can expand the volume size via the AWS CLI or the AWS EC2 Console:

# Modify volume size to 100 GiB (replace vol-0123456789abcdef0 with your volume ID)
aws ec2 modify-volume --volume-id vol-0123456789abcdef0 --size 100

Wait until the volume state transitions from modifying to in-use.

---

Step 2: Verify Block Device Visibility on the Linux Host

SSH into your Linux EC2 instance and inspect block devices:

# Display block devices and partition layout
lsblk

Expected output:

NAME         MAJ:MIN RM  SIZE RO TYPE MOUNTPOINTS
nvme0n1      259:0    0  100G  0 disk 
├─nvme0n1p1  259:1    0   40G  0 part /
└─nvme0n1p128 259:2   0    1M  0 part 

Notice that the physical disk (nvme0n1) reflects 100G, but partition 1 (nvme0n1p1) is still 40G.

---

Step 3: Install and Run growpart to Extend the Partition

Install growpart if not already present on your AMI:

# On Ubuntu / Debian:
apt-get update && apt-get install -y cloud-guest-utils

# On AlmaLinux / Rocky Linux / RHEL / Amazon Linux 2023:
dnf install -y cloud-utils-growpart

Grow the target partition (note the space between the device path and partition index):

# For NVMe devices (e.g. nvme0n1 partition 1):
growpart /dev/nvme0n1 1

# For Xen block devices (e.g. xvda partition 1):
growpart /dev/xvda 1

Expected output:

CHANGED: partition=1 start=2048 old: size=83884032 end=83886080 new: size=209713119 end=209715167

---

Step 4: Resize the Linux Filesystem

Identify your filesystem type:

# Inspect filesystem type and mount point
df -Th

Option A: For XFS Filesystems (Default in RHEL, AlmaLinux, Rocky Linux, Amazon Linux)

# Run xfs_growfs passing the mount point
xfs_growfs -d /

Option B: For ext4 Filesystems (Default in Ubuntu and Debian)

# Run resize2fs passing the partition device path
resize2fs /dev/nvme0n1p1

---

Troubleshooting

Problem: “growpart: NOCHANGE: partition 1 is size 83884032. it cannot be grown”

Possible cause: The EBS volume modification has not yet propagated to the virtual hypervisor, or you specified an incorrect partition index.

Check:

lsblk

Solution: Run partprobe or wait 60 seconds for the AWS hypervisor to update the virtual NVMe controller.

Problem: “xfs_growfs: / is not a mounted XFS filesystem”

Possible cause: The root filesystem is formatted as ext4 instead of xfs.

Check:

df -Th /

Solution: Use resize2fs /dev/DEVICE_PARTITION instead of xfs_growfs.

---

Verify the Configuration

Verify that the operating system recognizes the expanded capacity:

df -h /

Expected output:

Filesystem      Size  Used Avail Use% Mounted on
/dev/nvme0n1p1   99G   14G   85G  15% /

---

Production Checklist

• Captured an EBS snapshot prior to partition expansion.
• Modified volume size in AWS Console / CLI.
• Confirmed disk size in lsblk.
• Grown partition boundary using growpart.
• Expanded filesystem using xfs_growfs or resize2fs.
• Verified available disk space with df -h.

---

Frequently Asked Questions

Can I decrease or shrink an EBS volume size online?

No. AWS EBS does not support shrinking volumes directly. To reduce volume size, you must attach a smaller new EBS volume and copy data over using rsync.

Will resizing an EBS volume cause downtime or dropped TCP connections?

No. Elastic Volumes and online filesystem resizing with xfs_growfs / resize2fs operate completely live with zero downtime.

---

Related Guides & Services

• How to Troubleshoot High Load on Linux Servers
• AWS Cloud Infrastructure Management
• Linux Server Administration & Tuning

---

Need Help Managing AWS Cloud Infrastructure?

If you need proactive cloud monitoring, automated EBS volume scaling, or multi-region disaster recovery architecture, connect with our AWS certified engineers.

Contact Us