Este conteúdo não está disponível no idioma selecionado.
Chapter 4. Creating bootc compatible base disk images with bootc-image-builder
The bootc-image-builder
, available as a Technology Preview, is a containerized tool to create disk images from bootc images. You can use the images that you build to deploy disk images in different environments, such as the edge, server, and clouds.
4.1. Introducing image mode for RHEL for bootc-image-builder
With the bootc-image-builder
tool, you can convert bootc images into disk images for a variety of different platforms and formats. Converting bootc images into disk images is equivalent to installing a bootc. After you deploy these disk images to the target environment, you can update them directly from the container registry.
The bootc-image-builder
can only pull and use images from public container repositories. Building base disk images which come from private registries by using bootc-image-builder
is not supported in this release. If your container image is stored in a private repository, bootc-image-builder
cannot pull the image because it is not able to authenticate to the registry. If you need to use an image from a private repository, you must authenticate to the registry first and then pull the container image before you use it with bootc-image-builder
. After pulling the image, you can run the bootc-image-builder command using the --local
option.
The bootc-image-builder tool
supports generating the following image types:
- Disk image formats, such as ISO, suitable for disconnected installations.
Virtual disk images formats, such as:
- QEMU copy-on-write (QCOW2)
- Amazon Machine Image (AMI)/ — Raw
- Virtual Machine Image (VMI)
Deploying from a container image is beneficial when you run VMs or servers because you can achieve the same installation result. That consistency extends across multiple different image types and platforms when you build them from the same container image. Consequently, you can minimize the effort in maintaining operating system images across platforms. You can also update systems that you deploy from these disk images by using the bootc
tool, instead of re-creating and uploading new disk images with bootc-image-builder
.
Generic base container images do not include any default passwords or SSH keys. Also, the disk images that you create by using the bootc-image-builder
tool do not contain the tools that are available in common disk images, such as cloud-init
. These disk images are transformed container images only.
Although you can deploy a rhel-9-bootc
image directly, you can also create your own customized images that are derived from this bootc image. The bootc-image-builder
tool takes the rhel-9-bootc
OCI container image as an input.
Additional resources
4.2. Installing bootc-image-builder
The bootc-image-builder
is intended to be used as a container and it is not available as an RPM package in RHEL. To access it, follow the procedure.
Prerequisites
-
The
container-tools
meta-package is installed. The meta-package contains all container tools, such as Podman, Buildah, and Skopeo. -
You are authenticated to
registry.redhat.io
. For details, see Red Hat Container Registry Authentication.
Procedure
Login to authenticate to
registry.redhat.io
:$ sudo podman login registry.redhat.io
Install the
bootc-image-builder
tool:$ sudo podman pull registry.redhat.io/rhel9/bootc-image-builder
Verification
List all images pulled to your local system:
$ sudo podman images REPOSITORY TAG IMAGE ID CREATED SIZE registry.redhat.io/rhel9/bootc-image-builder latest b361f3e845ea 24 hours ago 676 MB
Additional resources
4.3. Creating QCOW2 images by using bootc-image-builder
Build a RHEL bootc image into a QEMU Disk Images (QCOW2) image for the architecture that you are running the commands on.
The RHEL base image does not include a default user. Optionally, you can inject a user configuration with the --config
option to run the bootc-image-builder container. Alternatively, you can configure the base image with cloud-init
to inject users and SSH keys on first boot. See Users and groups configuration - Injecting users and SSH keys by using cloud-init.
Prerequisites
- You have Podman installed on your host machine.
-
You have
virt-install
installed on your host machine. -
You have root access to run the
bootc-image-builder
tool, and run the containers in--privileged
mode, to build the images.
Procedure
Optional: Create a
config.toml
to configure user access, for example:[[customizations.user]] name = "user" password = "pass" key = "ssh-rsa AAA ... user@email.com" groups = ["wheel"]
Run
bootc-image-builder
. Optionally, if you want to use user access configuration, pass theconfig.toml
as an argument.NoteIf you do not have the container storage mount and
--local
image options, your image must be public.The following is an example of creating a public QCOW2 image:
$ sudo podman run \ --rm \ -it \ --privileged \ --pull=newer \ --security-opt label=type:unconfined_t \ -v ./config.toml:/config.toml \ -v ./output:/output \ registry.redhat.io/rhel9/bootc-image-builder:latest \ --type qcow2 \ --config /config.toml \ quay.io/<namespace>/<image>:<tag>
The following is an example of creating a private QCOW2 image:
$ sudo podman run \ --rm \ -it \ --privileged \ --pull=newer \ --security-opt label=type:unconfined_t \ -v $(pwd)/config.toml:/config.toml:ro \ -v $(pwd)/output:/output \ -v /var/lib/containers/storage:/var/lib/containers/storage \ registry.redhat.io/rhel9/bootc-image-builder:latest \ --local --type qcow2 \ quay.io/<namespace>/<image>:<tag>
You can find the
.qcow2
image in the output folder.
Next steps
- You can deploy your image. See Deploying a container image using KVM with a QCOW2 disk image.
- You can make updates to the image and push the changes to a registry. See Managing RHEL bootc images.
4.4. Creating VMDK images by using bootc-image-builder
Create a Virtual Machine Disk (VMDK) from a bootc image and use it within VMware’s virtualization platforms, such as vSphere, or use the Virtual Machine Disk (VMDK) in VirtualBox.
Prerequisites
- You have Podman installed on your host machine.
-
You have authenticated to the Red Hat Registry by using the
podman login registry.redhat.io
. -
You have pulled the
rhel9/bootc-image-builder
container image.
Procedure
Create a
Containerfile
with the following content:FROM registry.redhat.io/rhel9/rhel-bootc:9.4 RUN dnf -y install cloud-init open-vm-tools && \ ln -s ../cloud-init.target /usr/lib/systemd/system/default.target.wants && \ rm -rf /var/{cache,log} /var/lib/{dnf,rhsm} && \ systemctl enable vmtoolsd.service
Build the bootc image:
# podman build . -t localhost/rhel-bootc-vmdk
Create a VMDK file from the previously created bootc image:
NoteIf you do not have the container storage mount and --local image options, your image must be public.
The following is an example of creating a public VMDK image:
# podman run \ --rm \ -it \ --privileged \ -v /var/lib/containers/storage:/var/lib/containers/storage \ -v ./output:/output \ --security-opt label=type:unconfined_t \ --pull newer \ registry.redhat.io/rhel9/bootc-image-builder:9.4 --local \ --type vmdk \ quay.io/<namespace>/<image>:<tag>
The following is an example of creating a private VMDK image:
# podman run \ --rm \ -it \ --privileged \ --pull=newer \ --security-opt label=type:unconfined_t \ -v $(pwd)/config.toml:/config.toml:ro \ -v $(pwd)/output:/output \ -v /var/lib/containers/storage:/var/lib/containers/storage \ registry.redhat.io/rhel9/bootc-image-builder:latest \ --local --type vmdk \ quay.io/<namespace>/<image>:<tag>
The
--local
option uses the local container storage to source the originating image to produce the VMDK instead of a remote repository.
A VMDK disk file for the bootc image is stored in the output/vmdk
directory.
Next steps
- You can make updates to the image and push the changes to a registry. See Managing RHEL bootc images.
4.5. Creating GCE images by using bootc-image-builder
Build a RHEL bootc image into a gce image for the architecture that you are running the commands on. The RHEL base image does not include a default user. Optionally, you can inject a user configuration with the --config
option to run the bootc-image-builder container. Alternatively, you can configure the base image with cloud-init
to inject users and SSH keys on first boot. See Users and groups configuration - Injecting users and SSH keys by using cloud-init.
Prerequisites
- You have Podman installed on your host machine.
-
You have root access to run the
bootc-image-builder
tool, and run the containers in--privileged
mode, to build the images.
Procedure
Optional: Create a
config.toml
to configure user access, for example:[[customizations.user]] name = "user" password = "pass" key = "ssh-rsa AAA ... user@email.com" groups = ["wheel"]
Run
bootc-image-builder
. Optionally, if you want to use user access configuration, pass theconfig.toml
as an argument.NoteIf you do not have the container storage mount and
--local
image options, your image must be public.The following is an example of creating a
gce
image:$ sudo podman run \ --rm \ -it \ --privileged \ --pull=newer \ --security-opt label=type:unconfined_t \ -v ./config.toml:/config.toml \ -v ./output:/output \ registry.redhat.io/rhel9/bootc-image-builder:latest \ --type gce \ --config /config.toml \ quay.io/<namespace>/<image>:<tag>
The following is an example of creating a private
gce
image:$ sudo podman run \ --rm \ -it \ --privileged \ --pull=newer \ --security-opt label=type:unconfined_t \ -v $(pwd)/config.toml:/config.toml:ro \ -v $(pwd)/output:/output \ -v /var/lib/containers/storage:/var/lib/containers/storage \ registry.redhat.io/rhel9/bootc-image-builder:latest \ --local --type gce \ quay.io/<namespace>/<image>:<tag>
You can find the
gce
image in the output folder.
Next steps
- You can deploy your image. See Deploying a container image using KVM with a QCOW2 disk image.
- You can make updates to the image and push the changes to a registry. See Managing RHEL bootc images.
4.6. Creating AMI images by using bootc-image-builder and uploading it to AWS
Create an Amazon Machine Image (AMI) from a bootc image and use it to launch an Amazon Web Service EC2 (Amazon Elastic Compute Cloud) instance.
Prerequisites
- You have Podman installed on your host machine.
-
You have an existing
AWS S3
bucket within your AWS account. -
You have root access to run the
bootc-image-builder
tool, and run the containers in--privileged
mode, to build the images. -
You have the
vmimport
service role configured on your account to import an AMI into your AWS account.
Procedure
Create a disk image from the bootc image.
- Configure the user details in the Containerfile. Make sure that you assign it with sudo access.
- Build a customized operating system image with the configured user from the Containerfile. It creates a default user with passwordless sudo access.
Optional: Configure the machine image with
cloud-init
. See Users and groups configuration - Injecting users and SSH keys by using cloud-init. The following is an example:FROM registry.redhat.io/rhel9/rhel-bootc:9.4 RUN dnf -y install cloud-init && \ ln -s ../cloud-init.target /usr/lib/systemd/system/default.target.wants && \ rm -rf /var/{cache,log} /var/lib/{dnf,rhsm}
NoteYou can also use
cloud-init
to add users and additional configuration by using instance metadata.Build the bootc image. For example, to deploy the image to an
x86_64
AWS machine, use the following commands:$ podman build -t quay.io/<namespace>/<image>:<tag> . $ podman push quay.io/<namespace>/<image>:<tag> .
Use the
bootc-image-builder
tool to create an AMI from the bootc container image.NoteIf you do not have the container storage mount and
--local
image options, your image must be public.The following is an example of creating a public AMI image:
$ sudo podman run \ --rm \ -it \ --privileged \ --pull=newer \ -v $HOME/.aws:/root/.aws:ro \ --env AWS_PROFILE=default \ registry.redhat.io/rhel9/bootc-image-builder:latest \ --type ami \ --aws-ami-name rhel-bootc-x86 \ --aws-bucket rhel-bootc-bucket \ --aws-region us-east-1 \ quay.io/<namespace>/<image>:<tag>
The following is an example of creating a private AMI image:
$ sudo podman run \ --rm \ -it \ --privileged \ --pull=newer \ -v $HOME/.aws:/root/.aws:ro \ --env AWS_PROFILE=default \ registry.redhat.io/rhel9/bootc-image-builder:latest \ --type ami \ --aws-ami-name rhel-bootc-x86 \ --aws-bucket rhel-bootc-bucket \ --local quay.io/<namespace>/<image>:<tag>
You can also inject all your AWS configuration parameters by using
--env AWS_*
.ImportantThe following flags must be specified all together. If you do not specify any flag, the AMI is exported to your output directory.
-
--aws-ami-name
- The name of the AMI image in AWS -
--aws-bucket
- The target S3 bucket name for intermediate storage when you are creating the AMI --aws-region
- The target region for AWS uploadsThe
bootc-image-builder
tool builds an AMI image and uploads it to your AWS s3 bucket by using your AWS credentials to push and register an AMI image after building it.
-
Next steps
- You can deploy your image. See Deploying a container image to AWS with an AMI disk image.
- You can make updates to the image and push the changes to a registry. See Managing RHEL bootc images.
Additional resources
4.7. Creating raw disk images by using bootc-image-builder
You can convert a bootc image to a Raw image with an MBR or GPT partition table by using bootc-image-builder
. The RHEL base image does not include a default user, so optionally, you can inject a user configuration with the --config
option to run the bootc-image-builder
container. Alternatively, you can configure the base image with cloud-init
to inject users and SSH keys on first boot. See Users and groups configuration - Injecting users and SSH keys by using cloud-init.
Prerequisites
- You have Podman installed on your host machine.
-
You have root access to run the
bootc-image-builder
tool, and run the containers in--privileged
mode, to build the images. - You have pulled your target container image in the container storage.
Procedure
Optional: Create a
config.toml
to configure user access, for example:[[customizations.user]] name = "user" password = "pass" key = "ssh-rsa AAA ... user@email.com" groups = ["wheel"]
Run
bootc-image-builder
. If you want to use user access configuration, pass theconfig.toml
as an argument:NoteIf you do not have the container storage mount and
--local
image options, your image must be public.The following is an example of creating a public RAW image:
$ sudo podman run \ --rm \ -it \ --privileged \ --pull=newer \ --security-opt label=type:unconfined_t \ -v /var/lib/containers/storage:/var/lib/containers/storage \ -v ./config.toml:/config.toml \ -v ./output:/output \ registry.redhat.io/rhel9/bootc-image-builder:latest \ --local \ --type raw \ --config /config.toml \ quay.io/<namespace>/<image>:<tag>
The following is an example of creating a private RAW image:
$ sudo podman run \ --rm \ -it \ --privileged \ --pull=newer \ --security-opt label=type:unconfined_t \ -v $(pwd)/config.toml:/config.toml:ro \ -v $(pwd)/output:/output \ -v /var/lib/containers/storage:/var/lib/containers/storage \ registry.redhat.io/rhel9/bootc-image-builder:latest \ --local --type raw \ quay.io/<namespace>/<image>:<tag>
You can find the
.raw
image in the output folder.
Next steps
- You can deploy your image. See Deploying a container image by using KVM with a QCOW2 disk image.
- You can make updates to the image and push the changes to a registry. See Managing RHEL bootc images.
4.8. Creating ISO images by using bootc-image-builder
You can use bootc-image-builder
to create an ISO from which you can perform an offline deployment of a bootable container.
Prerequisites
- You have Podman installed on your host machine.
-
You have root access to run the
bootc-image-builder
tool, and run the containers in--privileged
mode, to build the images.
Procedure
Optional: Create a
config.toml
to configure user access, for example:[[customizations.user]] name = "user" password = "pass" key = "ssh-rsa AAA ... user@email.com" groups = ["wheel"]
Run
bootc-image-builder
. If you do not want to add any configuration, omit the-v $(pwd)/config.toml:/config.toml
argument.NoteIf you do not have the container storage mount and --local image options, your image must be public.
The following is an example of creating a public ISO image:
$ sudo podman run \ --rm \ -it \ --privileged \ --pull=newer \ --security-opt label=type:unconfined_t \ -v /var/lib/containers/storage:/var/lib/containers/storage \ -v $(pwd)/config.toml:/config.toml \ -v $(pwd)/output:/output \ registry.redhat.io/rhel9/bootc-image-builder:latest \ --type iso \ --config /config.toml \ quay.io/<namespace>/<image>:<tag>
The following is an example of creating a private ISO image:
$ sudo podman run \ --rm \ -it \ --privileged \ --pull=newer \ --security-opt label=type:unconfined_t \ -v $(pwd)/config.toml:/config.toml:ro \ -v $(pwd)/output:/output \ -v /var/lib/containers/storage:/var/lib/containers/storage \ registry.redhat.io/rhel9/bootc-image-builder:latest \ --local --type iso \ quay.io/<namespace>/<image>:<tag>
You can find the
.iso
image in the output folder.
Next steps
You can use the ISO image on unattended installation methods, such as USB sticks or Install-on-boot. The installable boot ISO contains a configured Kickstart file. See Deploying a container image by using Anaconda and Kickstart.
WarningBooting the ISO on a machine with an existing operating system or data can be destructive, because the Kickstart is configured to automatically reformat the first disk on the system.
- You can make updates to the image and push the changes to a registry. See Managing RHEL bootable images.
4.9. Using bootc-image-builder to build ISO images with a Kickstart file
You can use a Kickstart file to configure various parts of the installation process, such as setting up users, customizing partitioning, and adding an SSH key. You can include the Kickstart file in an ISO build to configure any part of the installation process, except the deployment of the base image. For ISOs with bootc container base images, you can use a Kickstart file to configure anything except the ostreecontainer
command.
For example, you can use a Kickstart to perform either a partial installation, a full installation, or even omit the user creation. Use bootc-image-builder
to build an ISO image that contains the custom Kickstart to configure your installation process.
Prerequisites
- You have Podman installed on your host machine.
-
You have root access to run the
bootc-image-builder
tool, and run the containers in--privileged
mode, to build the images.
Procedure
Create your Kickstart file. The following Kickstart file is an example of a fully unattended Kickstart file configuration that contains user creation, and partition instructions.
[customizations.installer.kickstart] contents = """ lang en_GB.UTF-8 keyboard uk timezone CET user --name <user> --password <password> --plaintext --groups <groups> sshkey --username <user> ssh-<type> <public key> rootpw --lock zerombr clearpart --all --initlabel autopart --type=plain reboot --eject """
-
Save the Kickstart configuration in the
toml
format to inject the Kickstart content. For example,config.toml
. Run
bootc-image-builder
, and include the Kickstart file configuration that you want to add to the ISO build. Thebootc-image-builder
automatically adds theostreecontainer
command that installs the container image.NoteIf you do not have the container storage mount and
--local
image options, your image must be public.The following is an example of creating a public ISO image:
$ sudo podman run \ --rm \ -it \ --privileged \ --pull=newer \ --security-opt label=type:unconfined_t \ -v /var/lib/containers/storage:/var/lib/containers/storage \ -v $(pwd)/config.toml:/config.toml \ -v $(pwd)/output:/output \ registry.redhat.io/rhel9/bootc-image-builder:latest \ --type iso \ --config /config.toml \ quay.io/<namespace>/<image>:<tag>
The following is an example of creating a private ISO image:
$ sudo podman run \ --rm \ -it \ --privileged \ --pull=newer \ --security-opt label=type:unconfined_t \ -v $(pwd)/config.toml:/config.toml:ro \ -v $(pwd)/output:/output \ -v /var/lib/containers/storage:/var/lib/containers/storage \ registry.redhat.io/rhel9/bootc-image-builder:latest \ --local --type iso \ quay.io/<namespace>/<image>:<tag>
You can find the
.iso
image in the output folder.
Next steps
You can use the ISO image on unattended installation methods, such as USB sticks or Install-on-boot. The installable boot ISO contains a configured Kickstart file. See Deploying a container image by using Anaconda and Kickstart.
WarningBooting the ISO on a machine with an existing operating system or data can be destructive, because the Kickstart is configured to automatically reformat the first disk on the system.
- You can make updates to the image and push the changes to a registry. See Managing RHEL bootable images.
4.10. Verification and troubleshooting
- If you have any issues configuring the requirements for your AWS image, see the following documentation
- For more details on users, groups, SSH keys, and secrets, see