Red Hat OpenShift
Red Hat’s Kubernetes platform: Kubernetes plus its own OS, operators that install and upgrade everything, and stricter security defaults, sold by subscription.
- What it is
- Container orchestration
- Used for
- Platform as a service (PaaS)
- Part of
- Kubernetes distribution
- Type
- Commercial product
- Made by
- Red Hat
- License
- Proprietary subscription (built from open source; the OKD community distribution is Apache 2.0)
- Facts checked
- 6 October 2026
What it is
OpenShift Container Platform is a Kubernetes distribution that also owns the layers around it. Every node runs Red Hat Enterprise Linux CoreOS (RHCOS), an image-based, mostly read-only RHEL with the CRI-O runtime, which the cluster updates as one atomic unit through its Machine Config Operator. The platform itself (Kubernetes core services, networking, monitoring, the console, the image registry) is installed and upgraded by operators, which also update Kubernetes and the RHCOS hosts during an over-the-air upgrade. Operator Lifecycle Manager and the OperatorHub catalog install and update further operators, such as databases or a service mesh.
On top of Kubernetes it adds its own APIs and tools: Projects, Routes that publish a Service at a hostname through an HAProxy-based ingress router, a built-in image registry with ImageStreams, Source-to-Image builds, a web console and the oc CLI. It is also stricter by default: the restricted-v2 security context constraint (SCC) drops all Linux capabilities, blocks privilege escalation and runs each container as an arbitrary non-root UID from its project’s range, so images that assume root often need changes.
You can run it yourself, or as a managed service: Red Hat OpenShift Service on AWS (ROSA), operated jointly by AWS and Red Hat and billed through AWS, or Azure Red Hat OpenShift (ARO), operated jointly with Microsoft. OKD is the free community distribution built from the same code, now on CentOS Stream CoreOS instead of RHCOS.
Why teams pick it
- Batteries included: registry, builds, routes, monitoring, console and operator catalog arrive installed and tested together
- Over-the-air upgrades for the whole stack, from the node OS to Kubernetes to platform services, driven by operators
- Secure defaults (non-root arbitrary UIDs, dropped capabilities, SELinux) that many hand-built clusters lack
- One platform self-managed, on ROSA or on ARO, with Red Hat support and an 18-month lifecycle per minor release (longer with Extended Update Support)
Watch out for
- Paid: support, certified content and updates come with a Red Hat subscription; OKD is free but has no Red Hat support
- Opinionated and large: over 100 operator-managed components, and RHCOS is the only supported OS for any node
- Strict SCCs break images that expect to run as root or own their files; they must support an arbitrary UID in the root group
- OpenShift-only APIs such as Routes, ImageStreams and Projects can tie manifests and habits to OpenShift
More of the Kubernetes family
What all container orchestration have in common →
MicroK8sOpen sourceDistributionsCanonical’s Kubernetes in one snap package: installs with one command on Linux, forms a highly available cluster by itself at three nodes, and adds features as addons.
RKE2Open sourceDistributionsSUSE’s security-focused Kubernetes distribution: one binary per node, hardened defaults, FIPS-validated crypto and close alignment with upstream.
Talos LinuxOpen sourceDistributionsA Linux distribution that only runs Kubernetes: immutable, minimal, with no shell or SSH, and managed entirely through an API.
Amazon EKSAWSManaged servicesAWS runs the Kubernetes control plane for you; you choose how much of the worker side to hand over too.
kindOpen sourceFor local developmentKubernetes IN Docker: each cluster node is a container, so a conformant multi-node cluster starts and disappears inside a CI job.
minikubeOpen sourceFor local developmentA local Kubernetes cluster for developers on Linux, macOS or Windows, in a container, a VM or on the host itself, with laptop-friendly extras.
AgonesOpen sourceBuilt on itRuns dedicated multiplayer game servers on Kubernetes: keeps a fleet of them warm, hands one to each match, and never kills a server while players are on it.
KlutchOpen sourceBuilt on itA young open-source control plane, built on Crossplane, that lets many application clusters order databases from one central Kubernetes cluster.How it fits with other technologies
Create a project, run a pod and publish it with a Route · bash
oc new-project hello-openshift
oc create -f https://raw.githubusercontent.com/openshift/origin/master/examples/hello-openshift/hello-pod.json
oc expose pod/hello-openshift # creates a Service
oc expose svc hello-openshift # creates a Route with a hostname on the router
oc get routesSee it in a real stack
No architecture uses Red Hat OpenShift yet. Drop it into the playground and wire it up yourself.Open playground →Where it came from2010–2026
OpenShift predates Kubernetes. Red Hat bought Makara, a platform-as-a-service start-up, in late 2010 and launched OpenShift in May 2011 as a hosted PaaS for Ruby, Python, PHP and Java apps. In April 2012 it open-sourced the code as OpenShift Origin under Apache 2.0. Versions 1 and 2 already ran apps in Linux containers, but with Red Hat’s own container runtime and orchestration, in units Red Hat called gears and cartridges.
In 2013 Red Hat bet on Docker for the container format, and in early 2014 it looked for an orchestrator, evaluating Docker’s own efforts, Mesos and a home-grown design. In March 2014 Google showed Red Hat engineers the project that became Kubernetes. Red Hat rebuilt OpenShift on Docker and Kubernetes: Origin 1.0, the code behind OpenShift 3, shipped in June 2015.
In January 2018 Red Hat agreed to buy CoreOS for $250 million, gaining its Container Linux OS, the Tectonic Kubernetes platform and the Quay registry. OpenShift 4, announced in May 2019, combined them: RHCOS replaced general-purpose RHEL nodes, and operators now installed and upgraded the whole platform. The same month Microsoft and Red Hat launched ARO, and AWS and Red Hat followed with ROSA in March 2021.
Why it’s built this way
- Adopt Kubernetes instead of keeping its own orchestrator
- Red Hat compared Docker’s orchestration, Mesos and building its own. It chose Kubernetes for its primitives (pods, services, replication controllers, a real networking model), for Google’s experience with Borg, and for its fast-growing community.
- Make the OS part of the cluster
- OpenShift 4 delivers RHCOS as an immutable host image installed as a component of the cluster. The same operator-driven upgrade then covers the host and Kubernetes, and an OS update that goes wrong rolls back with a single reboot.
- Operators for the platform itself
- OpenShift is many containerized components. Encoding how to install and upgrade each one in an operator turned installs and upgrades of the platform, Kubernetes included, into an automated, over-the-air process.
- Arbitrary non-root user IDs by default
- Running every container as a random UID from the project’s range, not root, limits what a process can do if it escapes the container through a runtime bug. The cost is that images must work as any user.
Timeline
- 2010
Red Hat acquires Makara, the PaaS start-up behind the first OpenShift
- 2011
OpenShift launches in May as a hosted PaaS with its own container system
- 2012
Code released as OpenShift Origin under Apache 2.0 in April
- 2015
Origin 1.0, the code base of OpenShift 3, rebuilt on Docker and Kubernetes (June)
- 2018
Red Hat agrees to buy CoreOS (January); Origin renamed OKD (August)
- 2019
OpenShift 4 with RHCOS and operator-driven upgrades; Azure Red Hat OpenShift GA
- 2021
Red Hat OpenShift Service on AWS (ROSA) GA in March
- 2026
OKD 5.0 ships on CentOS Stream CoreOS 10 and Kubernetes 1.36
- OpenShift blog: Announcing OpenShift, the PaaS for developers who love open source (May 2011) · archived copy, original
- Red Hat blog: Announcing OpenShift Origin, open source code for PaaS (Apr 2012) · archived copy, original
- Red Hat blog, Joe Fernandes: Why Red Hat chose Kubernetes for OpenShift (Nov 2016)
- OpenShift Origin v1.0.0 release on GitHub (Jun 2015)
- Red Hat blog: OKD, renaming of OpenShift Origin with the 3.10 release (Aug 2018)
- Red Hat press release: Red Hat to acquire CoreOS (Jan 2018)
- Red Hat blog: Introducing Red Hat OpenShift 4 (May 2019)
- Red Hat press release: Red Hat and Microsoft announce general availability of Azure Red Hat OpenShift (May 2019)
- AWS News Blog: Red Hat OpenShift Service on AWS now generally available (Mar 2021)
- OKD 5.0.0-okd-scos.0 release on GitHub (Sep 2026)
- Wikipedia: OpenShift (Makara, gears and cartridges in v2)
- Red Hat docs: Creating images, support arbitrary user IDs (OpenShift 4.19) · archived copy, original
Sources14 · facts checked 6 October 2026
- Red Hat docs: Red Hat Enterprise Linux CoreOS (OpenShift 4.19) · archived copy, original
- Red Hat docs: Understanding Operators, OLM and OperatorHub (OpenShift 4.19) · archived copy, original
- Red Hat docs: Managing security context constraints (OpenShift 4.19) · archived copy, original
- Red Hat docs: Routes, in Ingress and load balancing (OpenShift 4.19) · archived copy, original
- Red Hat docs: OpenShift image registry overview (OpenShift 4.19) · archived copy, original
- Red Hat docs: Creating images (S2I, arbitrary user IDs) (OpenShift 4.19) · archived copy, original
- Red Hat: OpenShift Container Platform life cycle policy (18 months per minor, EUS add-ons)
- Red Hat: OpenShift product page (editions, ROSA, ARO)
- AWS docs: What is Red Hat OpenShift Service on AWS? (jointly supported and operated, Red Hat SRE, 99.95% SLA)
- AWS News Blog: ROSA now generally available, billed by AWS (Mar 2021)
- Red Hat press release: Azure Red Hat OpenShift, jointly managed by Microsoft and Red Hat (May 2019)
- Red Hat blog: Introducing Red Hat OpenShift 4 (operators, RHCOS, over-the-air updates, Prometheus and Grafana)
- OKD website: an opinionated Kubernetes with over 100 operator-managed components
- OKD 5.0.0-okd-scos.0 release on GitHub (CentOS Stream CoreOS 10, Kubernetes 1.36)