mlrun-fips

packaged by Chainguard

Chainguard Container for mlrun-fips

Open source MLOps orchestration framework and platform (FIPS-enabled)

Chainguard Containers are regularly-updated, secure-by-default container images.

Download this Container Image

For those with access, this container image is available on cgr.dev:

docker pull cgr.dev/ORGANIZATION/mlrun-fips:latest

Be sure to replace the ORGANIZATION placeholder with the name used for your organization's private repository within the Chainguard Registry.

Compatibility Notes

The Chainguard mlrun-fips and mlrun-api-fips images are FIPS-enabled, drop-in replacements for the upstream mlrun/mlrun and mlrun/mlrun-api images. Their runtime configuration mirrors the non-FIPS mlrun and mlrun-api images verbatim; only the crypto path differs.

FIPS Support

Chainguard's mlrun-fips images ship with a validated redistribution of the OpenSSL FIPS provider module. For more on FIPS support in Chainguard container images, consult the guide on FIPS-enabled Chainguard Containers on Chainguard Academy.

Both images report FIPS mode through the system OpenSSL provider, so you can confirm it at runtime:

docker run --rm cgr.dev/ORGANIZATION/mlrun-fips:latest python3 -c \
  "from cryptography.hazmat.backends.openssl import backend; print('FIPS:', backend._fips_enabled)"

mlrun-fips (SDK/CLI base):

• No entrypoint; the default command is python3, matching upstream.
• Runs as root under working directory /mlrun, with OpenMPI allowed to run as root (OMPI_ALLOW_RUN_AS_ROOT=1, OMPI_ALLOW_RUN_AS_ROOT_CONFIRM=1).
• Ships the mlrun CLI and the mlrun[complete] SDK stack.

mlrun-api-fips (control-plane API server):

• Entrypoint tini --, command bash /mlrun/start_api.sh, working directory /mlrun.
• Serves the HTTP API on port 8080 and declares /mlrun/db as a volume for its SQLite state.
• Preconfigured with the upstream environment (MLRUN_HTTPDB__*, MLRUN_IS_API_SERVER=true, UVICORN_PORT=8080, and so on).

Getting Started

Run the API server (it defaults to an embedded SQLite database, so no external services are required):

docker run -d --name mlrun-api -p 8080:8080 cgr.dev/ORGANIZATION/mlrun-api-fips:latest
curl -sf http://localhost:8080/api/v1/projects

Drive it from the SDK/CLI base image by pointing MLRUN_DBPATH at the API server:

docker run --rm -e MLRUN_DBPATH="http://mlrun-api:8080" \
  cgr.dev/ORGANIZATION/mlrun-fips:latest mlrun version

Deploying with Helm

These images are typically deployed together via the upstream mlrun Helm chart. The chart deploys the mlrun-api image as the control plane; override it in your values.yaml:

api:
  image:
    repository: cgr.dev/ORGANIZATION/mlrun-api-fips
    tag: latest

The mlrun-fips base image is not a deployed service. It is used two ways: as the SDK and CLI image you use to interact with the platform (pulled directly or run in a notebook), and as the default image that function and job pods run on when you submit work without specifying one. That default is the upstream mlrun/mlrun, which mlrun selects through its default_base_image setting rather than a chart value.

To run functions on the Chainguard FIPS base instead, set the image globally on the API server through the chart:

api:
  extraEnvKeyValue:
    MLRUN_DEFAULT_BASE_IMAGE: cgr.dev/ORGANIZATION/mlrun-fips:latest

Alternatively, set it per function in your SDK code:

project.set_function("job.py", name="train", image="cgr.dev/ORGANIZATION/mlrun-fips:latest")

default_base_image is also the image mlrun builds on top of during .deploy(), where package installs run against it. A minimal image without a shell or package manager can break those builds, so set it globally only when your functions run on the base image directly; otherwise set it per function, or use a -dev variant as the build base.

Documentation and Resources

• mlrun documentation
• mlrun GitHub repository
• mlrun Helm charts
• Chainguard Academy

What are Chainguard Containers?

Chainguard's free tier of Starter container images are built with Wolfi, our minimal Linux undistro.

All other Chainguard Containers are built with Chainguard OS, Chainguard's minimal Linux operating system designed to produce container images that meet the requirements of a more secure software supply chain.

The main features of Chainguard Containers include:

• Minimal design, without unnecessary software bloat
• Daily builds to ensure container images are up-to-date with available security patches
• High quality build-time SBOMs attesting to the provenance of all artifacts within the image
• Verifiable signatures provided by Sigstore
• Reproducible builds with Cosign and apko (read more about reproducibility)

For cases where you need container images with shells and package managers to build or debug, most Chainguard Containers come paired with a development, or -dev, variant.

In all other cases, including Chainguard Containers tagged as :latest or with a specific version number, the container images include only an open-source application and its runtime dependencies. These minimal container images typically do not contain a shell or package manager.

Although the -dev container image variants have similar security features as their more minimal versions, they include additional software that is typically not necessary in production environments. We recommend using multi-stage builds to copy artifacts from the -dev variant into a more minimal production image.

Need additional packages?

To improve security, Chainguard Containers include only essential dependencies. Need more packages? Chainguard customers can use Custom Assembly to add packages, either through the Console, chainctl, or API.

To use Custom Assembly in the Chainguard Console: navigate to the image you'd like to customize in your Organization's list of images, and click on the Customize image button at the top of the page.

Learn More

Refer to our Chainguard Containers documentation on Chainguard Academy. Chainguard also offers VMs and Libraries — contact us for access.

Trademarks

This software listing is packaged by Chainguard. The trademarks set forth in this offering are owned by their respective companies, and use of them does not imply any affiliation, sponsorship, or endorsement by such companies.