CRC for local OpenShift 4.x development clusters
CRC is a tool to help you run containers. It manages local VMs to run a OpenShift 4.x cluster, Microshift or Podman optimized for testing and development purposes
At a glance
- What is it?
- CRC manages a local virtual machine for an OpenShift 4.x or OKD cluster intended primarily for development and testing. It is a practical choice when developers need an OpenShift environment without the version or deployment complexity of older tools such as Minishift or CDK.
- Who is it for?
- Developers who need a disposable or repeatable OpenShift 4.x, OKD, Microshift, or Podman-oriented test environment should evaluate CRC. Teams needing OpenShift 3.x should reject it because the project explicitly directs them to older alternatives.
- Can I use it commercially?
- Yes. Apache-2.0 is a permissive licence: you can use, modify and sell software built on it, as long as you keep its copyright and licence notices.
- Is it still maintained?
- Yes. The repository last received commits 2 days ago.
- What is it written in?
- Mainly Go, according to GitHub's language statistics.
Answers come from the project's GitHub data, last synced on October 20, 2026, and from our analysis. They are not legal advice.
Editorial analysis
CRC's OpenShift 4.x scope and intended users
CRC addresses a specific gap between a normal container runtime and a full OpenShift development environment. The repository describes it as a tool that runs containers by managing a local virtual machine containing an OpenShift 4.x cluster. It also supports OKD, and the project description identifies Microshift and Podman as additional use cases optimized for testing and development. This places CRC closer to a managed development cluster than to a general-purpose replacement for `docker` or `podman`.
The primary audience is developers and testers who need OpenShift APIs, operators, deployment behavior, or other cluster-level facilities on a workstation. A lightweight container engine can run individual images, but that does not by itself establish an OpenShift 4.x control plane. CRC handles the surrounding local VM lifecycle so the developer can concentrate on the cluster rather than assembling the virtual machine and operating-system configuration independently.
CRC is not intended to reproduce every deployment model. The README explicitly says its cluster is optimized for testing and development, not presented as a production installation. It also excludes OpenShift 3.x. For that generation, the README points to `oc cluster up`, Minishift, or CDK. That boundary matters because choosing CRC for a legacy requirement would begin with a version mismatch, regardless of how convenient the CLI appears.
The local VM and cluster lifecycle managed by CRC
Based on the repository layout and README, CRC is a Go application with command-line code under `cmd/`, package implementation under `pkg/`, packaging under `packaging/`, and integration-oriented assets under `test/`. The command and package separation is consistent with a CLI that coordinates machine lifecycle, configuration, and cluster variants rather than embedding an OpenShift implementation in the project itself.
The documented operating boundary is clear: CRC manages a local VM, and the VM carries the selected cluster environment. Published binaries can include an embedded OpenShift disk image, which reduces the need for a user to locate and attach an image separately. The Makefile identifies supported platform variables such as `GOOS` and `GOARCH`, along with separate OpenShift, OKD, and Microshift versions. Those build settings show that a release can associate its CRC binary with particular cluster image versions.
The README does not document the full startup sequence, network configuration, storage layout, DNS behavior, or stop and delete lifecycle in enough detail to claim how those internals work. It also does not specify whether a particular binary permits runtime switching among every cluster type. That omission matters for adoption because knowing that CRC starts an OpenShift VM does not answer every operational question about cluster replacement or host networking.
Downloading CRC and making the first cluster start
The README does not provide a universal package-manager installation command. It directs users to the CRC binaries page for downloads, and states that the available binaries can contain an embedded OpenShift disk image. The appropriate package or image should therefore be selected from that page for the workstation before any CLI operation is attempted.
CRC asks for telemetry consent the first time it runs. A user can set that preference explicitly before the next start with the documented command:
crc config set consent-telemetry <yes/no>Replace `<yes/no>` with `yes` or `no` according to the local policy. The README says approved collection includes pseudonymized usage data sent to Red Hat servers and links the privacy statement and data table for the specific collection scope. After setting consent, the documented next operation is to start CRC:
crc startThe README links to the Getting Started Guide at `crc.dev/docs` for the complete setup flow, but it does not enumerate expected host packages, resources, or post-start console output. Those checks belong in the linked guide rather than an invented short command sequence.
OpenShift, OKD, Microshift, and Podman are different CRC targets
The repository metadata presents CRC as a manager for OpenShift 4.x, Microshift, or Podman, while the README specifically documents OpenShift 4.x and OKD. These names should not be treated as interchangeable cluster modes. OpenShift and OKD are the documented VM-based cluster targets in the README, while the build system separately records `OPENSHIFT_VERSION`, `OKD_VERSION`, and `MICROSHIFT_VERSION`.
That distinction has practical consequences. A developer validating manifests against an OpenShift API, RBAC behavior, or cluster operators needs a target that actually provides those services. A developer who specifically requires Podman behavior or a lighter runtime should verify which downloadable artifact and configuration provide that target. The README does not show a single command for selecting among them, so switching by assumption could result in an unsupported or unexpected setup.
The Makefile currently associates version 2.64.0 with OpenShift 4.22.14, OKD 4.22.0-okd-scos.10, and Microshift 4.22.13. These are repository build settings, not a promise that every released binary contains every image. The README specifically describes OpenShift binaries with an embedded disk image. Check the download page and Getting Started Guide for the exact artifact and target before using commands beyond the documented telemetry preference.
Where CRC is the wrong development tool
The clearest wrong-tool case is OpenShift 3.x. CRC does not solve that requirement, and the README directs users to `oc cluster up`, Minishift, or CDK instead. Another mismatch occurs when the goal is simply to run one container with no OpenShift control plane. In that case, CRC introduces a VM-managed cluster platform where a direct container runtime would be a closer fit.
CRC may also be unsuitable when production-like infrastructure, multi-node topology, or operational guarantees are required. The README describes a single local development and testing environment, not a production cluster manager. It gives no evidence about production support, high availability, disaster recovery, or multi-node scheduling. It would therefore be unsafe to infer production suitability from the software managing an OpenShift VM successfully on a laptop.
The documentation also leaves host preparation largely outside the README. Virtualization support, available memory, disk capacity, and compatible host configuration may affect first start, but no thresholds or host-specific setup instructions appear in the README. The linked Getting Started Guide is the place to establish those prerequisites. CRC should be rejected for a workstation that cannot satisfy them or when a cloud-hosted, shared, or production environment is the actual requirement.
CRC compared with Minishift, CDK, and container engines
CRC differs from Minishift and CDK mainly in the platform generation it targets. The README identifies all three as alternatives for OpenShift 3.x, while CRC manages OpenShift 4.x or OKD. For a project that still depends on OpenShift 3.x APIs or tooling, that version boundary is more important than differences in convenience or installer presentation.
CRC also differs from an ordinary container engine in scope. Podman is named by the repository description as a supported development context, and the Makefile uses `CONTAINER_RUNTIME ?= podman` for its build configuration. Running a container through Podman does not mean using CRC for that workload. CRC is relevant when the workload needs the surrounding environment it manages, such as an OpenShift or OKD cluster VM. A project that only needs image execution, volume mounts, and local container commands has a simpler path through the runtime alone.
For OpenShift 3.x developers, the project itself supplies the practical comparison: use `oc cluster up`, Minishift, or CDK. For OpenShift 4.x development and testing, CRC's documented focus is narrower and more direct. The decision should follow the required OpenShift generation and environment type, not a general preference between local cluster tools.
Maintenance, upgrades, telemetry, and Apache-2.0 licensing
The repository is not archived, and its last push was on 2026-10-15. Recent releases include v2.64.0 on 2026-09-24, v2.63.0 on 2026-08-18, and v2.62.0 on 2026-07-03. This indicates recent project activity, although the documentation does not define a support duration or guarantee compatibility with every host platform.
Upgrading CRC may involve more than replacing a small CLI binary because the Makefile associates releases with specific OpenShift, OKD, and Microshift versions. The documentation does not describe whether disk images are replaced in place, whether an existing VM can be retained, or what cleanup preserves application data. The Getting Started Guide or release-specific documentation must settle those details before an upgrade plan is considered complete.
Telemetry is another upgrade-relevant setting. Approval allows pseudonymized usage data to be sent to Red Hat servers, and the preference can be changed with `crc config set consent-telemetry <yes/no>` before the next `crc start`. The repository is licensed under Apache-2.0. That license permits use and modification under its terms, but organizations still need to review those terms themselves, including notices and redistribution conditions.
Editorial conclusion
Developers who need a disposable or repeatable OpenShift 4.x, OKD, Microshift, or Podman-oriented test environment should evaluate CRC. Teams needing OpenShift 3.x should reject it because the project explicitly directs them to older alternatives. Before adoption, verify that the downloaded binary matches the host, confirm the intended OpenShift or OKD disk image, and review the initial telemetry consent because that choice can be changed before the next CRC start.
Frequently asked questions
How do I install CRC for a local OpenShift cluster?
The README does not give a package-manager command. It directs users to https://console.redhat.com/openshift/create/local for CRC binaries, some of which include an embedded OpenShift disk image.
Can CRC run OpenShift 3.x?
No. CRC targets OpenShift 4.x and OKD. For OpenShift 3.x, the README points to `oc cluster up`, Minishift, or CDK.
How can I disable telemetry before starting CRC?
Run `crc config set consent-telemetry no` before the next `crc start`. CRC otherwise asks for telemetry consent the first time it runs.
Official sources
Add this badge to your README
If you maintain this project, the badge below links readers to this analysis and shows its maintenance status from the daily GitHub snapshot. Paste the markdown into your README; add ?metric=license or ?metric=stars to the image URL for a different field.
[](https://hysenlabs.com/projects/crc-org-crc)