Configuring a High-Availability Storage with GlusterFS on Ubuntu

glusterfs

network

filesystem

Ubuntu

Reviewed on May 02, 2025

GlusterFS is an open-source, scalable network file system suited for high data-intensive workloads, such as media streaming, cloud storage, and content delivery networks (CDN). In this tutorial, we will deploy a high-availability storage solution using GlusterFS 11 and Scaleway Block Storage on Ubuntu 22.04.

Each storage Instance will mirror the other, and files will automatically be replicated across all servers.

Before you start

To complete the actions presented below, you must have:

A Scaleway account logged into the console
Owner status or IAM permissions allowing you to perform actions in the intended Organization
An SSH key
Three Instances running Ubuntu 22.04 LTS
Attached Scaleway Block Storage to each GlusterFS Instance

Configuring the host file

Before installing GlusterFS, ensure each Instance can resolve the others via their hostname. This will simplify network communication between servers.

Log in to each Instance via SSH.
```
ssh root@<your_instance_public_ip>
```
Update the system:
```
apt update && apt upgrade -y
```
Edit the /etc/hosts file on each Instance:
```
nano /etc/hosts
```

Add entries for each Instance:

<IP_ADDRESS_1> gluster01
<IP_ADDRESS_2> gluster02
<IP_ADDRESS_3> client01

Verify hostname resolution:

ping -c 3 gluster01
ping -c 3 gluster02
ping -c 3 client01

Adding the GlusterFS 11 repository

Install the software-properties-common package on all Instances:
```
apt install software-properties-common -y
```

Add the GlusterFS 11 repository to all systems:

sudo add-apt-repository ppa:gluster/glusterfs-11
apt update

Installing GlusterFS server

Install the GlusterFS server on both gluster01 and gluster02:
```
apt install glusterfs-server -y
```

Start the glusterd service and enable it at boot:

systemctl start glusterd.service
systemctl enable glusterd.service

Verify the installation:
```
systemctl status glusterd.service
glusterfs --version
```
You should see GlusterFS 11.x and an active status for the service.

Setting up a distributed GlusterFS volume with Scaleway Block Storage

Important

In a production environment, it is strongly recommended to use dedicated storage, such as Scaleway Block Storage, instead of system directories for GlusterFS volumes. This ensures better data safety, scalability, and performance. Using system directories (like /glusterfs) can lead to potential issues if the root partition fills up.

Attaching Scaleway Block Storage

Create and attach Scaleway Block Storage to each Instance:
- From the Scaleway Console, create a new Block Storage volume for each GlusterFS server.
- Attach each volume to the respective Instance.
- Once attached, log into each Instance and check if the Block Storage is recognized:
```
lsblk
```
  You should see a new unmounted device (e.g., /dev/sdb).
Partition the volume and format it with a file system (e.g., ext4):
```
mkfs.ext4 /dev/sdb
```
Mount the Block Storage to a directory:
- Create a mount point for GlusterFS:
```
mkdir -p /mnt/glusterfs
```
- Mount the volume to the directory:
```
mount /dev/sdb /mnt/glusterfs
```
Add the Block Storage to /etc/fstab to ensure it mounts automatically after reboot:
```
/dev/sdb /mnt/glusterfs ext4 defaults 0 0
```
Repeat these steps and attach, format, and mount a Block Storage volume on all GlusterFS Instances (e.g., gluster01 and gluster02).

Creating a distributed GlusterFS volume

Now that Block Storage is mounted on both servers, proceed with creating the distributed GlusterFS volume:

Create a GlusterFS Volume using the mounted block storage directories on both servers. Run the following command on gluster01:
```
gluster volume create vol01 transport tcp gluster01:/mnt/glusterfs gluster02:/mnt/glusterfs force
```
Start the volume:
```
gluster volume start vol01
```
Check the volume status:
```
gluster volume info
```

Configuring the GlusterFS client

Install the GlusterFS client on client01:
```
apt install glusterfs-client -y
```
Create a mount point:
```
mkdir -p /mnt/glusterfs
```

Mount the GlusterFS volume:

mount -t glusterfs gluster01:/vol01 /mnt/glusterfs

Verify the mount:
```
df -h /mnt/glusterfs
```

Add the volume to /etc/fstab for permanent mounting:

gluster01:/vol01 /mnt/glusterfs glusterfs defaults,_netdev 0 0

Testing replication and mirroring

Mount the GlusterFS volume on both gluster01 and gluster02:

mount -t glusterfs gluster01:/vol01 /mnt
mount -t glusterfs gluster02:/vol01 /mnt

From client01, create test files in /mnt/glusterfs:
```
touch file01 file02 file03
```
Verify that the files are replicated on both gluster01 and gluster02:
```
ls /mnt
```

You should see the same files on both servers, confirming the replication is working.

Questions?

Visit our Help Center and find the answers to your most frequent questions.

Visit Help Center