Use cloud-init to configure a swap partition on a Linux VM

Applies to: ✔️ Linux VMs ✔️ Flexible scale sets

This article shows you how to use cloud-init to configure the swap partition on various Linux distributions. The swap partition was traditionally configured by the Linux Agent (WALA) based on which distributions required one. This document outlines the process for building the swap partition on demand during provisioning time using cloud-init. For more information about how cloud-init works natively in Azure and the supported Linux distros, see cloud-init overview

By default on Azure, Ubuntu gallery images don't create swap partitions. To enable swap partition configuration during VM provisioning time using cloud-init - please see the AzureSwapPartitions document on the Ubuntu wiki.

Create a file in your current shell named cloud_init_swappart.txt and paste the following configuration. For this example, create the file on your local machine. You can use any editor you wish. Make sure that the whole cloud-init file is copied correctly, especially the first line.

#cloud-config
disk_setup:
  ephemeral0:
    table_type: gpt
    layout: [66, [33,82]]
    overwrite: true
fs_setup:
  - device: ephemeral0.1
    filesystem: ext4
  - device: ephemeral0.2
    filesystem: swap
mounts:
  - ["ephemeral0.1", "/mnt"]
  - ["ephemeral0.2", "none", "swap", "sw,nofail,x-systemd.requires=cloud-init.service", "0", "0"]

The mount is created with the nofail option to ensure that the boot process continues even if the mount isn't completed successfully.

Before deploying this image, you need to create a resource group with the az group create command. An Azure resource group is a logical container into which Azure resources are deployed and managed. The following example creates a resource group named myResourceGroup in the chinaeast location.

az group create --name myResourceGroup --location chinaeast

Now, create a VM with az vm create and specify the cloud-init file with --custom-data cloud_init_swappart.txt as follows:

az vm create \
  --resource-group myResourceGroup \
  --name vmName \
  --image imageCIURN \
  --custom-data cloud_init_swappart.txt \
  --generate-ssh-keys

If you already provisioned your server and wish to modify the mount point of the ephemeral storage and want to configure a part of the disk as swap space, use the following steps.

Create cloud-init configuration file named 00-azure-swap.cfg in the /etc/cloud/cloud.cfg.d directory with the following content:

#cloud-config
disk_setup:
  ephemeral0:
    table_type: gpt
    layout: [66, [33,82]]
    overwrite: true
fs_setup:
  - device: ephemeral0.1
    filesystem: ext4
  - device: ephemeral0.2
    filesystem: swap
mounts:
  - ["ephemeral0.1", "/mnt"]
  - ["ephemeral0.2", "none", "swap", "sw,nofail,x-systemd.requires=cloud-init.service", "0", "0"]

Next, append a line to the /etc/systemd/system.conf file with following content:

DefaultEnvironment="CLOUD_CFG=/etc/cloud/cloud.cfg.d/00-azure-swap.cfg"

After the changes are done, the machine needs to be deallocated or redeployed for the changes to take effect.

SSH to the public IP address of your VM shown in the output from the preceding command. Enter your own user and publicIpAddress as follows:

ssh <user>@<publicIpAddress>

Once you have SSH'ed into the vm, check if the swap partition was created

sudo swapon -s

The output from this command should look like this:

Filename                Type        Size    Used    Priority
/dev/sdb2  partition   2494440 0   -1

For more cloud-init examples of configuration changes, see the following:

• Add an additional Linux user to a VM
• Run a package manager to update existing packages on first boot
• Change VM local hostname
• Install an application package, update configuration files and inject keys