Learn about the platform

Red Hat OpenStack Services on OpenShift (RHOSO) provides the foundation to build a private or public Infrastructure-as-a-Service (IaaS) cloud on top of Red Hat Enterprise Linux. It is a scalable, potentially fault-tolerant platform for the development of cloud-enabled workloads.

The RHOSO control plane is hosted and managed as a workload on a Red Hat OpenShift Container Platform (RHOCP) cluster. The RHOSO data plane consists of external Red Hat Enterprise Linux (RHEL) nodes that host RHOSO workloads. The RHEL nodes are managed on the data plane with an automation execution environment. The data plane nodes can be Compute nodes, Storage nodes, Networker nodes, or other types of nodes.

The RHOSO IaaS cloud is implemented by a collection of interacting services that control its computing, storage, and networking resources. You can manage the cloud with a web-based interface to control, provision, and automate RHOSO resources. Additionally, an extensive API controls the RHOSO infrastructure and this API is also available to end users of the cloud.

Note
RHOSO only supports RHOCP master and worker nodes with processors based on a 64-bit x86 hardware architecture.

Environment features
Copy link

RHOSO provides container-native deployment with services on OpenShift and RHEL, supporting both installer- and user-provisioned infrastructure.

Container-native application delivery
RHOSO is delivered by using a container-native approach that spans the Red Hat OpenShift Container Platform (RHOCP) and RHEL platforms to deliver a container-native RHOSO deployment.
RHOCP-hosted services
RHOCP hosts infrastructure services and RHOSO controller services by using RHOCP Operators to provide lifecycle management.
Ansible-managed RHEL-hosted services
RHOSO workloads run on RHEL nodes that are managed by the OpenStack Operator. The OpenStack Operator runs Ansible jobs to configure the RHEL data plane nodes, such as the Compute nodes. RHOCP manages provisioning, DNS, and configuration management.
Installer-provisioned infrastructure
The RHOSO installer enables installer-provisioned infrastructure that uses RHOSO bare-metal machine management to provision the Compute nodes for the RHOSO cloud.
User-provisioned infrastructure
If you have your own machine ingest and provisioning workflow, you can use the RHOSO pre-provisioned model to add your pre-provisioned hardware into your RHOSO environment, while receiving the benefits of a container-native workflow.
Hosted RHOSO client
RHOSO provides a host openstackclient pod that is preconfigured with administrator access to the deployed RHOSO environment.

Service operators
Copy link

RHOSO IaaS services run as Operators on an OpenShift cluster to manage compute, storage, networking, and other cloud services.

Important

You use the Red Hat OpenShift Container Platform (RHOCP) OperatorHub to obtain and create the OpenStack Operator.

The OpenStack Operator (openstack-operator) installs and deploys all the service Operators detailed in the Service Operators table, and is the interface that you use to manage those Operators.

Table 1. Service Operators
Service Operator Default Description

Bare Metal Provisioning (ironic)

ironic-operator

Disabled

Supports physical machines for a variety of hardware vendors with hardware-specific drivers. Bare Metal Provisioning integrates with the Compute service to provision physical machines in the same way that virtual machines are provisioned, and provides a solution for the bare-metal-to-trusted-project use case.

openstack-baremetal-operator

Enabled

Used by the OpenStack Operator during the bare-metal node provisioning process.

Block Storage (cinder)

cinder-operator

Enabled

Provides and manages persistent block storage volumes for virtual machine instances.

Compute (nova)

nova-operator

Enabled

Provides management of the provisioning of compute resources, such as Virtual Machines, through the libvirt driver or physical servers through the ironic driver.

Dashboard (horizon)

horizon-operator

Disabled

Provides a browser-based GUI dashboard for creating and managing cloud resources and user access. The Dashboard service provides Project, Admin, and Settings dashboards by default. You can configure the dashboard to interface with other products such as billing, monitoring, and additional management tools.

DNS (designate)

designate-operator

Disabled

Provides DNS as a service that manages DNS records, names, and zones in the cloud. The DNS service provides a REST API, and is integrated with the RHOSO Networking service (neutron) to automatically create records for virtual machine instances, network ports, and floating IPs. User management is available from integration with the Identity service (keystone).

Identity (keystone)

keystone-operator

Enabled

Provides user authentication in the cloud and authorization to all RHOSO services and for managing users, projects, and roles. Supports multiple authentication mechanisms, including username and password credentials, token-based systems, and AWS-style log-ins.

Image (glance)

glance-operator

Enabled

Registry service for storing resources such as virtual machine images and volume snapshots. Cloud users can add new images or take a snapshot of an existing instance for immediate storage. You can use the snapshots for backup or as templates for new instances.

Key Management (barbican)

barbican-operator

Enabled

Provides secure storage, provisioning and management of secrets such as passwords, encryption keys, and X.509 Certificates. This includes keying material such as Symmetric Keys, Asymmetric Keys, Certificates, and raw binary data.

Load balancing (octavia)

octavia-operator

Disabled

Provides load balancing as a service for the cloud that supports multiple provider drivers. The reference provider driver is amphora, which is an open source, scalable, and highly available load balancing provider. It accomplishes its delivery of load balancing services by managing a fleet of virtual machines, collectively known as amphorae, which it creates on-demand.

MariaDB

mariadb-operator

Enabled

Provides methods to deploy and manage MariaDB Galera clusters.

Memcached

infra-operator

Enabled

Provides methods for managing infrastructure.

Networking (neutron)

neutron-operator

Enabled

Provides Networking-as-a-Service (NaaS) through software-defined networking (SDN) in virtual compute environments. Handles the creation and management of a virtual networking infrastructure in the cloud, which includes networks, subnets, and routers.

Object Storage (swift)

swift-operator

Enabled

Provides efficient and durable storage of large amounts of data, including static entities such as videos, images, email messages, files, or instance images. Objects are stored as binaries on the underlying file system with metadata stored in the extended attributes of each file.

OVN

ovn-operator

Enabled

Provides methods to deploy and manage OVNs.

Orchestration (heat)

heat-operator

Disabled

Template-based orchestration engine that supports automatic creation of resource stacks. Provides templates to create and manage cloud resources such as storage, networking, instances, or applications. You can use the templates to create stacks, which are collections of resources.

Placement (placement)

placement-operator

Enabled

Provides methods to install and manage an OpenStack Placement installation.

RabbitMQ

rabbitmq-cluster-operator

Enabled

Provides methods to deploy and manage RabbitMQ clusters.

Shared File Systems (manila)

manila-operator

Disabled

Provisions shared file systems that can be used by multiple virtual machine instances, bare-metal nodes, or containers.

Telemetry (ceilometer, prometheus)

telemetry-operator

Enabled

Provides user-level usage data for RHOSO clouds. You can use the data for customer billing, system monitoring, or alerts. Telemetry can collect data from notifications sent by existing RHOSO components such as Compute usage events, or by polling RHOSO infrastructure resources such as libvirt.

Test

test-operator

Disabled

A test Operator that automates the testing of a RHOSO deployment on a RHOCP cluster.

Limitations and unsupported features
Copy link

Review the unsupported features and known limitations for Red Hat OpenStack Services on OpenShift (RHOSO) to ensure your deployment meets your infrastructure requirements.

Block Storage service (cinder):

  • Block Storage service replication is not supported in RHOSO unless it is included in a certified partner’s disaster recovery (DR) solution.
  • The Logical Volume Manager (LVM) driver is not supported.
  • NFS versions earlier than 4 are not supported.

Compute service (nova):

  • Off-path Network Backends are not supported in RHOSO 18.0.
  • Customizing policies is not supported. If you require custom policies, contact Red Hat for a support exception.
  • The following packages are not supported in RHOSO:
    • nova-serialproxy
    • nova-spicehtml5proxy
  • File injection of personality files to inject user data into virtual machine instances is not supported. As a workaround, users can pass data to their instances by using the --user-data option to run a script during instance boot, or set instance metadata by using the --property option when launching an instance.
  • Persistent memory for instances (vPMEM) is not supported. You can create persistent memory namespaces only on Compute nodes that have NVDIMM hardware. Red Hat has removed support for persistent memory from RHOSP 17.0 and later in response to the announcement by the Intel Corporation on July 28, 2022 that they are discontinuing investment in their Intel® Optane™ business.
  • QEMU emulation of non-native architectures is not supported.
  • LVM is not supported as an image back end.
  • The ploop image format is not supported.
  • NFS versions earlier than 4 are not supported.

Image service (glance):

  • RHOSO supports only one architecture, x86_64. There is no valid use case that requires this to be set for a RHOSO cloud, so all hosts are x86_64.
  • NFS versions earlier than 4 are not supported.

Optimize service (watcher):

  • The Optimize service engine does not revert actions for failed Action Plans.
  • Limited data integration between the Optimize service and other services, such as the Compute service, can result in Action Plans that assign migration actions to unsuitable nodes.
  • When you create an audit with a strategy such as zone_migration or hardware_maintenance that specifies the intended destination or backup Compute node, the Optimize service does not validate your selection against Compute service constraints. This can lead to a failed migration action.
  • The creation and execution of Action Plans are asynchronous. The status of the cluster can change between these operations which can cause invalid Action Plans and a failed Watcher action.
  • When the cross-az feature is disabled in the Compute service, an instance cannot run in a different availability zone to the one that is used by its attached Block Storage service (cinder) volumes. Watcher does not validate this constraint during its audit process and might plan a migration to a host located in a different availability zone, which results in failure.
  • Do not run more than one instance of the Watcher decision-engine and applier services, because the architectural limitations of Watcher in RHOSO 18 can cause unfinished actions and audits being orphaned in scale-down scenarios.

Shared File Systems service (manila):

  • NFS versions earlier than 4.1 are not supported for CephFS-NFS back ends. Linux clients must use NFSv4.1 or later. NFSv3 is available for clients that do not yet support NFSv4.1, such as Microsoft Windows, but recovery is not expected for NFSv3 clients when a CephFS-NFS service fails over. Simultaneous access from Windows and Linux clients is not supported.