1. Documentation
×

Hosted control planes overview

You can deploy OKD clusters by using two different control plane configurations: standalone or hosted control planes. The standalone configuration uses dedicated virtual machines or physical machines to host the control plane. With hosted control planes for OKD, you create control planes as pods on a hosting cluster without the need for dedicated virtual or physical machines for each control plane.

Glossary of common concepts and personas for hosted control planes

When you use hosted control planes for OKD, it is important to understand its key concepts and the personas that are involved.

Concepts

hosted cluster

An OKD cluster with its control plane and API endpoint hosted on a management cluster. The hosted cluster includes the control plane and its corresponding data plane.

hosted cluster infrastructure

Network, compute, and storage resources that exist in the tenant or end-user cloud account.

hosted control plane

An OKD control plane that runs on the management cluster, which is exposed by the API endpoint of a hosted cluster. The components of a control plane include etcd, the Kubernetes API server, the Kubernetes controller manager, and a VPN.

hosting cluster

See management cluster.

managed cluster

A cluster that the hub cluster manages. This term is specific to the cluster lifecycle that the multicluster engine for Kubernetes Operator manages in Red Hat Advanced Cluster Management. A managed cluster is not the same thing as a management cluster. For more information, see Managed cluster.

management cluster

An OKD cluster where the HyperShift Operator is deployed and where the control planes for hosted clusters are hosted. The management cluster is synonymous with the hosting cluster.

management cluster infrastructure

Network, compute, and storage resources of the management cluster.

node pool

A resource that contains the compute nodes. The control plane contains node pools. The compute nodes run applications and workloads.

Personas

cluster instance administrator

Users who assume this role are the equivalent of administrators in standalone OKD. This user has the cluster-admin role in the provisioned cluster, but might not have power over when or how the cluster is updated or configured. This user might have read-only access to see some configuration projected into the cluster.

cluster instance user

Users who assume this role are the equivalent of developers in standalone OKD. This user does not have a view into OperatorHub or machines.

cluster service consumer

Users who assume this role can request control planes and worker nodes, drive updates, or modify externalized configurations. Typically, this user does not manage or access cloud credentials or infrastructure encryption keys. The cluster service consumer persona can request hosted clusters and interact with node pools. Users who assume this role have RBAC to create, read, update, or delete hosted clusters and node pools within a logical boundary.

cluster service provider

Users who assume this role typically have the cluster-admin role on the management cluster and have RBAC to monitor and own the availability of the HyperShift Operator as well as the control planes for the tenant’s hosted clusters. The cluster service provider persona is responsible for several activities, including the following examples:

  • Owning service-level objects for control plane availability, uptime, and stability

  • Configuring the cloud account for the management cluster to host control planes

  • Configuring the user-provisioned infrastructure, which includes the host awareness of available compute resources

Introduction to hosted control planes

You can use hosted control planes for Red Hat OKD to reduce management costs, optimize cluster deployment time, and separate management and workload concerns so that you can focus on your applications.

Hosted control planes is available by using the multicluster engine for Kubernetes Operator version 2.0 or later on the following platforms:

  • Bare metal by using the Agent provider

  • OKD Virtualization

  • Amazon Web Services (AWS), as a Technology Preview feature

  • IBM Z, as a Technology Preview feature

  • IBM Power, as a Technology Preview feature

Architecture of hosted control planes

OKD is often deployed in a coupled, or standalone, model, where a cluster consists of a control plane and a data plane. The control plane includes an API endpoint, a storage endpoint, a workload scheduler, and an actuator that ensures state. The data plane includes compute, storage, and networking where workloads and applications run.

The standalone control plane is hosted by a dedicated group of nodes, which can be physical or virtual, with a minimum number to ensure quorum. The network stack is shared. Administrator access to a cluster offers visibility into the cluster’s control plane, machine management APIs, and other components that contribute to the state of a cluster.

Although the standalone model works well, some situations require an architecture where the control plane and data plane are decoupled. In those cases, the data plane is on a separate network domain with a dedicated physical hosting environment. The control plane is hosted by using high-level primitives such as deployments and stateful sets that are native to Kubernetes. The control plane is treated as any other workload.

Diagram that compares the hosted control plane model against OpenShift with a coupled control plane and workers

Benefits of hosted control planes

With hosted control planes for OKD, you can pave the way for a true hybrid-cloud approach and enjoy several other benefits.

  • The security boundaries between management and workloads are stronger because the control plane is decoupled and hosted on a dedicated hosting service cluster. As a result, you are less likely to leak credentials for clusters to other users. Because infrastructure secret account management is also decoupled, cluster infrastructure administrators cannot accidentally delete control plane infrastructure.

  • With hosted control planes, you can run many control planes on fewer nodes. As a result, clusters are more affordable.

  • Because the control planes consist of pods that are launched on OKD, control planes start quickly. The same principles apply to control planes and workloads, such as monitoring, logging, and auto-scaling.

  • From an infrastructure perspective, you can push registries, HAProxy, cluster monitoring, storage nodes, and other infrastructure components to the tenant’s cloud provider account, isolating usage to the tenant.

  • From an operational perspective, multicluster management is more centralized, which results in fewer external factors that affect the cluster status and consistency. Site reliability engineers have a central place to debug issues and navigate to the cluster data plane, which can lead to shorter Time to Resolution (TTR) and greater productivity.

Differences between hosted control planes and OKD

Hosted control planes is a form factor of OKD. Hosted clusters and the stand-alone OKD clusters are configured and managed differently. See the following tables to understand the differences between OKD and hosted control planes:

Cluster creation and lifecycle

OKD Hosted control planes

You install a standalone OKD cluster by using the openshift-install binary or the Assisted Installer.

You install a hosted cluster by using the hypershift.openshift.io API resources such as HostedCluster and NodePool, on an existing OKD cluster.

Cluster configuration

OKD Hosted control planes

You configure cluster-scoped resources such as authentication, API server, and proxy by using the config.openshift.io API group.

You configure resources that impact the control plane in the HostedCluster resource.

etcd encryption

OKD Hosted control planes

You configure etcd encryption by using the APIServer resource with AES-GCM or AES-CBC. For more information, see "Enabling etcd encryption".

You configure etcd encryption by using the HostedCluster resource in the SecretEncryption field with AES-CBC or KMS for Amazon Web Services.

Operators and control plane

OKD Hosted control planes

A standalone OKD cluster contains separate Operators for each control plane component.

A hosted cluster contains a single Operator named Control Plane Operator that runs in the hosted control plane namespace on the management cluster.

etcd uses storage that is mounted on the control plane nodes. The etcd cluster Operator manages etcd.

etcd uses a persistent volume claim for storage and is managed by the Control Plane Operator.

The Ingress Operator, network related Operators, and {olm-first} run on the cluster.

The Ingress Operator, network related Operators, and {olm-first} run in the hosted control plane namespace on the management cluster.

The OAuth server runs inside the cluster and is exposed through a route in the cluster.

The OAuth server runs inside the control plane and is exposed through a route, node port, or load balancer on the management cluster.

Updates

OKD Hosted control planes

The Cluster Version Operator (CVO) orchestrates the update process and monitors the ClusterVersion resource. Administrators and OpenShift components can interact with the CVO through the ClusterVersion resource. The oc adm upgrade command results in a change to the ClusterVersion.Spec.DesiredUpdate field in the ClusterVersion resource.

The hosted control planes update results in a change to the .spec.release.image field in the HostedCluster and NodePools resources. Any changes to the ClusterVersion resource are ignored.

After you update an OKD cluster, both the control plane and compute machines are updated.

After you update the hosted cluster, only the control plane is updated. You perform node pool updates separately.

Machine configuration and management

OKD Hosted control planes

The MachineSets resource manages machines in the openshift-machine-api namespace.

The NodePool resource manages machines on the management cluster.

A set of control plane machines are available.

A set of control plane machines do not exist.

You enable a machine health check by using the MachineHealthCheck resource.

You enable a machine health check through the .spec.management.autoRepair field in the NodePool resource.

You enable autoscaling by using the ClusterAutoscaler and MachineAutoscaler resources.

You enable autoscaling through the spec.autoScaling field in the NodePool resource.

Machines and machine sets are exposed in the cluster.

Machines, machine sets, and machine deployments from upstream Cluster CAPI Operator are used to manage machines but are not exposed to the user.

All machine sets are upgraded automatically when you update the cluster.

You update your node pools independently from the hosted cluster updates.

Only an in-place upgrade is supported in the cluster.

Both replace and in-place upgrades are supported in the hosted cluster.

The Machine Config Operator manages configurations for machines.

The Machine Config Operator does not exist in hosted control planes.

You configure machine Ignition by using the MachineConfig, KubeletConfig, and ContainerRuntimeConfig resources that are selected from a MachineConfigPool selector.

You configure the MachineConfig, KubeletConfig, and ContainerRuntimeConfig resources through the config map referenced in the spec.config field of the NodePool resource.

The Machine Config Daemon (MCD) manages configuration changes and updates on each of the nodes.

For an in-place upgrade, the node pool controller creates a run-once pod that updates a machine based on your configuration.

You can modify the machine configuration resources such as the SR-IOV Operator.

You cannot modify the machine configuration resources.

Networking

OKD Hosted control planes

The Kube API server communicates with nodes directly, because the Kube API server and nodes exist in the same Virtual Private Cloud (VPC).

The Kube API server communicates with nodes through Konnectivity. The Kube API server and nodes exist in a different Virtual Private Cloud (VPC).

Nodes communicate with the Kube API server through the internal load balancer.

Nodes communicate with the Kube API server through an external load balancer or a node port.

Web console

OKD Hosted control planes

The web console shows the status of a control plane.

The web console does not show the status of a control plane.

You can update your cluster by using the web console.

You cannot update the hosted cluster by using the web console.

The web console displays the infrastructure resources such as machines.

The web console does not display the infrastructure resources.

You can configure machines through the MachineConfig resource by using the web console.

You cannot configure machines by using the web console.
Additional resources

Relationship between hosted control planes, multicluster engine Operator, and RHACM

You can configure hosted control planes by using the multicluster engine for Kubernetes Operator. The multicluster engine is an integral part of Red Hat Advanced Cluster Management (RHACM) and is enabled by default with RHACM. The multicluster engine Operator cluster lifecycle defines the process of creating, importing, managing, and destroying Kubernetes clusters across various infrastructure cloud providers, private clouds, and on-premises data centers.

The multicluster engine Operator is the cluster lifecycle Operator that provides cluster management capabilities for OKD and RHACM hub clusters. The multicluster engine Operator enhances cluster fleet management and supports OKD cluster lifecycle management across clouds and data centers.

Cluster life cycle and foundation
Figure 1. Cluster life cycle and foundation

You can use the multicluster engine Operator with OKD as a standalone cluster manager or as part of a RHACM hub cluster.

A management cluster is also known as the hosting cluster.

You can deploy OKD clusters by using two different control plane configurations: standalone or hosted control planes. The standalone configuration uses dedicated virtual machines or physical machines to host the control plane. With hosted control planes for OKD, you create control planes as pods on a management cluster without the need for dedicated virtual or physical machines for each control plane.

RHACM and the multicluster engine Operator introduction diagram
Figure 2. RHACM and the multicluster engine Operator introduction diagram

Versioning for hosted control planes

With each major, minor, or patch version release of OKD, two components of hosted control planes are released:

  • The HyperShift Operator

  • The hcp command-line interface (CLI)

The HyperShift Operator manages the lifecycle of hosted clusters that are represented by the HostedCluster API resources. The HyperShift Operator is released with each OKD release. The HyperShift Operator creates the supported-versions config map in the hypershift namespace. The config map contains the supported hosted cluster versions.

You can host different versions of control planes on the same management cluster.

Example supported-versions config map object
    apiVersion: v1
    data:
      supported-versions: '{"versions":["4.14"]}'
    kind: ConfigMap
    metadata:
      labels:
        hypershift.openshift.io/supported-versions: "true"
      name: supported-versions
      namespace: hypershift

You can use the hcp CLI to create hosted clusters.

You can use the hypershift.openshift.io API resources, such as, HostedCluster and NodePool, to create and manage OKD clusters at scale. A HostedCluster resource contains the control plane and common data plane configuration. When you create a HostedCluster resource, you have a fully functional control plane with no attached nodes. A NodePool resource is a scalable set of worker nodes that is attached to a HostedCluster resource.

The API version policy generally aligns with the policy for Kubernetes API versioning.

Additional resources