Code Monkey home page Code Monkey logo

knative-build's Introduction

Openshift Knative build

This repository holds Openshift's fork of knative/build with additions and fixes needed only for the OpenShift side of things.

List of releases

How this repository works ?

The master branch holds up-to-date specific openshift files that are necessary for CI setups and maintaining it. This includes:

  • Scripts to create a new release branch from upstream
  • CI setup files
    • operator configuration (for Openshift's CI setup)
    • tests scripts
  • Operator's base configurations

Each release branch holds the upstream code for that release and our openshift's specific files.

CI Setup

For the CI setup, two repositories are of importance:

  • This repository
  • openshift/release which contains the configuration of CI jobs that are run on this repository

All of the following is based on OpenShift’s CI operator configs. General understanding of that mechanism is assumed in the following documentation.

The job manifests for the CI jobs are generated automatically. The basic configuration lives in the /ci-operator/config/openshift/knative-build folder of the openshift/release repository. These files include which version to build against (OCP 4.0 for our recent cases), which images to build (this includes all the images needed to run Knative and also all the images required for running e2e tests) and which command to execute for the CI jobs to run (more on this later).

Before we can create the ci-operator configs mentioned above, we need to make sure there are Dockerfiles for all images that we need (they’ll be referenced by the ci-operator config hence we need to create them first). The generate-dockerfiles.sh script takes care of creating all the Dockerfiles needed automatically. The files now need to be committed to the branch that CI is being setup for.

The basic ci-operator configs mentioned above are generated via the generate-release.sh file in the openshift/knative-build repository. They are generated to alleviate the burden of having to add all possible test images to the manifest manually, which is error prone.

Once the file is generated, it must be committed to the openshift/release repository, as the other manifests linked above. The naming schema is openshift-knative-build-BRANCH.yaml, thus the files existing already correspond to our existing releases and the master branch itself.

After the file has been added to the folder as mentioned above, the job manifests itself will need to be generated as is described in the corresponding ci-operator documentation.

Once all of this is done (Dockerfiles committed, ci-operator config created and job manifests generated) a PR must be opened against openshift/release to include all the ci-operator related files. Once this PR is merged, the CI setup for that branch is active.

Create a new release

Deliverables:

  • Tagged images on quay.io
  • An OLM manifest referencing these images
  • Install documentation

High-level steps for a release

Building upstream

  1. Create a release branch from the upstream’s release tag, i.e. release-v0.5.0. This is created in the fork that we maintain of upstream. See our branching instructions for deeper information.
  2. Create a CI job for that branch in openshift/release. See our CI setup instructions for deeper information.
  3. Do whatever you need to do to make this CI pass
  4. Create a “dummy” PR with a ci file, which contains the current output of “date”. This is to trigger CI explicitly.
  5. Make sure that PR is green and merge it. This will trigger the images to become available on the CI registry.

Mirroring images to quay.io

  1. Make sure the images for the release are built and “promoted”. This can be verified by “docker pull”ing them, for example: “docker pull registry.svc.ci.openshift.org/openshift/knative-v0.5.0:knative-eventing-controller”
  2. If that’s the case, create/amend an image mirroring mapping file as described here.

Update the operator

  1. Copy the upstream release manifest to the operator’s deploy/resources directory. All files in this directory will be applied, so remove any old ones.
  2. Update the manifest[s] to replace gcr.io images with quay.io images.
  3. Build/test the operator
  4. Commit the source
  5. export VERSION="vX.Y.Z" # replace X.Y.Z as appropriate 6.. operator-sdk build quay.io/openshift-knative/$OPERATOR_NAME:$VERSION
  6. docker push quay.io/openshift-knative/$OPERATOR_NAME:$VERSION
  7. git tag $VERSION; git push --tags

Update OLM metadata

  1. The following instructions amount to mirroring upstream changes in the OLM manifests beneath openshift/olm in our forks.
  2. Create a new *.clusterserviceversion.yaml for the upstream release. It’s easiest to copy the previous release’s CSV to a new file and update the name, version, and replaces fields, as well as the version of the operator’s image.
  3. Ensure the upstream release’s RBAC policies match what’s in the CSV. Any upstream edits should be carried over.
  4. Mirror any upstream manifest changes to CRD’s in the corresponding *.crd.yaml file.
  5. Update the currentCSV field in the *.package.yaml file.
  6. Regenerate the *.catalogsource.yaml file using the catalog.sh script. For example,
DIR=openshift/olm \
~/src/knative-operators/etc/scripts/catalog.sh \
> openshift/olm/knative-build.catalogsource.yaml

Tag the repository

Get a "LGTM" from QE

Get a "LGTM" from Docs

Gather release notes from JIRA/GitHub

Send a release announcement

knative-build's People

Contributors

adrcunha avatar alanfx avatar arilivigni avatar bkrannich avatar bsnchan avatar dgageot avatar dprotaso avatar evankanderson avatar imjasonh avatar jcrossley3 avatar jonjohnsonjr avatar mattmoor avatar matzew avatar mdemirhan avatar mgencur avatar mvinkler avatar openshift-merge-robot avatar pmorie avatar rootfs avatar sebgoa avatar shashwathi avatar vdemeester avatar zrss avatar

Watchers

 avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.