Deploying a .img File on a Cloud Server via Rescue System

Introduction

This tutorial explains how to successfully deploy an .img file on a cloud server using the rescue system.
This process is required when installing a custom operating system image on the server.

Prerequisites

• A cloud server
  • Activated rescue system
  • SSH access to the server
• A volume to temporarily store the .img file (as the primary disk will be overwritten)
• An operating system .img file (or .qcow2, see step 3)
• Basic Linux command-line skills

Step 1 - Activating and Connecting to the Rescue System

• Activating the Rescue System

  1. Log in to the cloud provider's management console.
  2. Select the target server and activate the rescue system.
  3. If required, add an SSH key for easier authentication.
  4. Reboot the server to enter rescue mode.

• Connecting to the Server

  Make an SSH connection to the rescue system:

  ssh root@<your_server_ip>

  After logging in, the rescue system banner will appear.

Step 2 - Identifying the Target Disk and Mounting the Volume

Since the .img file cannot be stored directly on the primary disk, a volume is required as a storage location.
To list all available volumes, use the following command:

lsblk

Example output:

NAME    MAJ:MIN RM  SIZE RO TYPE MOUNTPOINTS
sda       8:0    0  76G  0 disk 
sdb       8:16   0  40G  0 disk 

Here, sda is the primary disk to be overwritten, while sdb is the volume to store the .img file.

Manually Mounting the Volume

In the rescue system, additional volumes are not mounted automatically.
Now create a mount directory and mount the volume:

mkdir -p /mnt/vol1
mount /dev/sdX /mnt/vol1

Replace /dev/sdX with the actual volume name (e.g., /dev/sdb).

If you are unsure which volume to use, check the partition table:

fdisk -l

Step 3 - Preparing the Image File

If the uploaded or downloaded image is in .qcow2 format, it needs to be converted:

qemu-img convert -f qcow2 -O raw /mnt/vol1/source.qcow2 /mnt/vol1/destination.img

With an image in .img format, you can choose from three methods:

• Method 1: Upload via SCP (Command Line)

  If the .img file is stored locally, it can be uploaded using SCP:

  scp /path/to/image.img root@<your_server_ip>:/mnt/vol1/

• Method 2: Upload via SFTP (FileZilla)

  For users who prefer a graphical interface, FileZilla can be used:

  1. Open FileZilla.

  2. Create a new connection:

     Option	Value
     Server	<your_server_ip>
     Username	root
     Passwort	<your_password>
     Port	22
     Connection Type	SFTP - SSH File Transfer Protocol

  3. Drag and drop the .img file into the /mnt/vol1/ directory.

• Method 3: Download using wget

  If the .img file is hosted externally, it can be downloaded directly to the server:

  wget <URL_to_image.img> -O /mnt/vol1/image.img

Step 4 - Writing the Image to the Target Disk

• Recommended Method: pv | dd

  Now, pv | dd is used for better visibility of progress:

  pv /mnt/vol1/image.img | dd of=/dev/sda bs=4M status=progress

  If pv is not installed, it can be added with:

  apt update && apt install -y pv

• Alternative: dd Directly

  If pv is not used, the image can be written directly with dd:

  dd if=/mnt/vol1/image.img of=/dev/sda bs=4M status=progress

Step 5 - Reboot and Check the System

Once the image has been successfully written, the server can be rebooted.

Use the following command to reboot into the installed operating system:

reboot

After the reboot, verify that the system has booted successfully and is accessible.

Troubleshooting Boot Problems

If the system does not boot correctly, consult the operating system documentation.
In some cases, a bootloader check may be required.

For general bootloader configuration, see:

• GNU GRUB - Official Documentation:
  https://www.gnu.org/software/grub/manual/grub/grub.html

• UEFI Specification and Documentation:
  https://uefi.org/specifications

Note on Alternative Image Formats

Some alternative image formats can also be converted and used.
However, there is a risk that they may not work properly due to driver or hardware compatibility issues.

Cloud servers are based on KVM virtualization, so ensure that the operating system you choose is KVM-compatible.

Conclusion

The custom operating system should now be installed on the server.

License: MIT