Skip to content

gursewak1997/fedora-coreos-pipeline

 
 

Repository files navigation

Fedora CoreOS Pipeline

This is the Jenkins pipeline configuration for Fedora CoreOS.

The pipeline is built around coreos-assembler.

It uses the OpenShift Jenkins template and is meant to be fully compatible with the local dev oc cluster up workflow. For more information on getting started, see HACKING.

The production instance is running in an OpenShift cluster in Fedora's infrastructure (though note anonymous view is blocked by default). Its raw build output can be seen in the build browser (but note that the latest supported version of FCOS must be downloaded from the official page).

To operate the production Jenkins (or more generally to access the production namespace), you must have access to the cluster at https://console-openshift-console.apps.ocp.fedoraproject.org/ and to the "fedora-coreos" project.

If you need access, you can open a pull request to https://pagure.io/fedora-infra/ansible/blob/main/f/playbooks/openshift-apps/fedora-coreos-pipeline.yml with your Fedora Account System username similar to this one.

You also need one of the project admins as a sponsor, please reach out on Libera.Chat #fedora-coreos channel.

Finally you will need to run the Ansible playbook from batcave.

sudo rbac-playbook -l os_control openshift-apps/fedora-coreos-pipeline.yml

You should also be able to run this pipeline and run it in any OpenShift cluster that supports (potentially nested) virtualization.

Terminology

This repo tries to maintain a consistent set of words to avoid confusion around different concepts with similar names:

So for example, a developer pipeline may perform e.g. a production or development stream build, but release tooling only cares about builds performed by the official pipeline pushed to the official locations.

Avoid using the naked word devel. Always either use development (if talking about the streams) or developer (if talking about the pipeline).

Similarly, avoid using the word production alone, in favour of production stream or official pipeline.

Releases

No releases published

Packages

No packages published

Languages

  • Groovy 74.0%
  • Python 12.7%
  • HCL 9.9%
  • Shell 3.4%