Preparing an Image for Deploying a Cloud Server

Before creating a cloud server from an image, you need to configure the image in advance. This ensures a successful installation and reliable operation of the control panel.

The following section covers the checks and settings required to deploy the server successfully.

Basic Settings

Partition Table

Make sure the disk has the correct partition structure. We recommend using MBR or GPT partitions, as they are supported by most Linux distributions and can be configured with command-line tools such as fdisk, gdisk and others.

gdisk -l /dev/vda

Sample output:

GPT fdisk (gdisk) version 1.0.5
Partition table scan:
  MBR: MBR only
  BSD: not present
  APM: not present
  GPT: not present

If a different partition structure is used, you may need to recreate the image with the correct configuration.

Disk Layout

If the disk has only one partition (the root partition) and it uses the ext3 or ext4 file system, the file system will be resized automatically when the server is installed from the image. It will be expanded to the disk size specified in the server configuration.

For example, if the block device in the image is 10 GB and the disk size specified at installation is 15 GB, the block device and file system will be automatically expanded to 15 GB.

The file system will not be expanded if LVM (Logical Volume Manager) is used.

OS Bootloader

We use SeaBIOS to run virtual machines, so a standard BIOS bootloader must be installed in the OS. If a UEFI bootloader is installed, remove it and install a BIOS bootloader.

If you use GRUB2, you can find installation and configuration instructions here.

Configuring /etc/fstab

To ensure stable system operation, configure the /etc/fstab file to mount the root partition by UUID. Example:

UUID=f19002a1-6e7a-45ac-91cd-24b7cc0e4cd9 / ext4 defaults 0 1

To get the UUID of a partition, you can use the command:

blkid

Sample output:

/dev/vda1: UUID="f19002a1-6e7a-45ac-91cd-24b7cc0e4cd9" TYPE="ext4" PARTUUID="f7a1fae1-01"
/dev/loop0: TYPE="squashfs"
/dev/loop1: TYPE="squashfs"
/dev/loop2: TYPE="squashfs"

Main User Account

When the server is created, a password will be generated for the root user (or Administrator on Windows). Make sure this user exists. The user normally exists unless it was deliberately deleted.

qemu-guest-agent

Check that the qemu-guest-agent service is installed and running. It is required for certain operations from the control panel, such as creating backups.

• Check the service status.

Systemd:

systemctl status qemu-guest-agent.service

OpenRC and init.d:

/etc/init.d/qemu-guest-agent status

or:

service qemu-guest-agent status

• Install the agent if it is not present on the system:

Debian/Ubuntu:

apt-get install qemu-guest-agent

CentOS / RHEL:

yum install qemu-guest-agent

Alpine Linux:

apk add qemu-guest-agent

• Enable it at startup:

Systemd:

systemctl enable qemu-guest-agent.service

init.d:

chkconfig --add qemu-guest-agent

OpenRC:

rc-update add qemu-guest-agent

On Windows, you need to install this service using a special ISO image, which you can download here. You can also use it to install the virtio drivers required to support virtualized devices.

These settings are sufficient to create a cloud server from the image. However, we recommend completing the remaining settings for more convenient server management.

Additional Settings

Network Settings

Once the server is running, it is important to make sure it is reachable on the network. To do this, configure the DHCP client to obtain IPv4 and IPv6 addresses automatically.

On most systems, a DHCP client is installed by default, but we recommend using the dhclient utility with the settings below:

option rfc3442-classless-static-routes code 121 = array of unsigned integer 8;
option dhcp6.next-hop code 242 = ip6-address;

send host-name = gethostname();
request subnet-mask, broadcast-address, time-offset, routers,
        domain-name, domain-name-servers, domain-search, host-name,
        dhcp6.name-servers, dhcp6.domain-search, dhcp6.fqdn, dhcp6.sntp-servers,
        netbios-name-servers, netbios-scope, interface-mtu,
        rfc3442-classless-static-routes, ntp-servers, dhcp6.next-hop;

timeout 300;

SSH Access

Since SSH is usually used for remote access to the server, it is important to check that SSH is installed and starts when the server boots.

• Check the status:

Systemd:

systemctl status ssh.service

OpenRC and init.d:

/etc/init.d/ssh status

or:

service ssh status

• Install SSH if it is not present:

Debian/Ubuntu:

apt-get install openssh-server

CentOS / RHEL:

yum -y install openssh-server openssh-clients

Alpine Linux:

apk add openssh

• Enable it at startup:

Systemd:

systemctl enable ssh.service

Init.d:

chkconfig --add ssh

OpenRC:

rc-update add sshd

• Also make sure that your firewall allows access to the SSH port (22 by default).

Disabling swap

We recommend disabling the swap file.

• Check the current state of the swap file:

swapon --show

Sample output:

NAME      TYPE SIZE USED PRIO
/swap.img file 685M   0B   -2

• Disable swap:

swapoff -a