Coder Social home page Coder Social logo

build-definitions's Introduction

build-definitions

This repository contains components that are installed or managed by the managed CI and Build Team.

This includes default Pipelines and Tasks. You need to have bootstrapped a working appstudio configuration from (see https://github.com/redhat-appstudio/infra-deployments) for the dev of pipelines or new tasks.

Pipelines and Tasks are delivered into App Studio via quay organization redhat-appstudio-tekton-catalog. Pipelines are bundled and pushed into repositories prefixed with pipeline- and tagged with $GIT_SHA (tag will be updated with every change). Tasks are bundled and pushed into repositories prefixed with task- and tagged with $VERSION where VERSION is the task version (tag is updated when the task file contains any change in the PR)

Currently a set of utilities are bundled with App Studio in quay.io/redhat-appstudio/appstudio-utils:$GIT_SHA as a convenience but tasks may be run from different per-task containers.

Building

Script hack/build-and-push.sh creates bundles for pipelines, tasks and create appstudio-utils image. Images are pushed into your quay.io repository. You will need to set MY_QUAY_USER to use this feature and be logged into quay.io on your workstation. Once you run the hack/build-and-push.sh all pipelines will come from your bundle instead of from the default installed by gitops into the cluster.

Note

If you're using Mac OS, you need to install GNU coreutils before running the hack/build-and-push.sh script:

brew install coreutils

There is an option to push all bundles to a single quay.io repository (this method is used in PR testing). It is used by setting TEST_REPO_NAME environment variable. Bundle names are then specified in the container image tag, i.e. quay.io/<quay-user>/$TEST_REPO_NAME:<bundle-name>-<tag>

Pipelines

The pipelines can be found in the pipelines directory.

  • core-services: contains pipeline for the CI of Stonesoup core services e.g. application-service and build-service.
  • template-build: contains common template used to generate docker-build, fbc-builder, java-builder and nodejs-builder pipelines

Tasks

The tasks can be found in the tasks directories. Tasks are bundled and used by bundled pipelines. Tasks are not stored in the Cluster. For quick local innerloop style task development, you may install new Tasks in your local namespace manually and create your pipelines as well as the base task image to test new function. Tasks can be installed into local namespace using oc apply -k tasks/appstudio-utils/util-tasks.

There is a container which is used to support multiple set of tasks called quay.io/redhat-appstudio/appstudio-utils:GIT_SHA , which is a single container which is used by multiple tasks. Tasks may also be in their own container as well however many simple tasks are utilities and will be packaged for app studio in a single container. Tasks can rely on other tasks in the system which are co-packed in a container allowing combined tasks (build-only vs build-deploy) which use the same core implementations.

Shellspec tests can be run by invoking hack/test-shellspec.sh.

Release

Release is done by setting env variable MY_QUAY_USER=redhat-appstudio-tekton-catalog, BUILD_TAG=$(git rev-parse HEAD) and running hack/build-and-push.sh.

Versioning

When the task update changes the interface (eg. change of parameters, workspaces or results names) then a new version of the task should be created. The folder with the new version must contain MIGRATION.md with instructions on how to update the current pipeline file in user's .tekton folder.

Adding a new parameter with a default value does not require the task version increase.

Task version increase must be approved by Project Manager.

Testing

Script ./hack/test-builds.sh creates pipelines and tasks directly in current namespace and executes there test builds. By setting the environment variable MY_QUAY_USER the images will be pushed into user's quay repository, in that case creation of secret named redhat-appstudio-staginguser-pull-secret is required.

Script ./hack/test-build.sh provides way to test on custom git repository and pipeline. Usage example: ./hack/test-build.sh https://github.com/jduimovich/spring-petclinic java-builder.

Compliance

Task definitions must comply to Enterprise Contract policies. Currently, there are two policy configurations. The all-tasks policy configuration applies to all Task definitions, while the build-tasks policy configuration applies only to build Task definitions. A build Task, i.e. one that produces a container image, must abide to both policy configurations.

build-definitions's People

Contributors

arewm avatar brunoapimentel avatar chmeliik avatar dirgim avatar gabemontero avatar gbenhaim avatar hongweiliu17 avatar jduimovich avatar jencull avatar jinqi7 avatar joejstuart avatar jsztuka avatar lcarva avatar martinbasti avatar michkov avatar mkosiarc avatar mmorhun avatar psturc avatar rcerven avatar renovate[bot] avatar rh-tap-build-team[bot] avatar robnester-rh avatar sbose78 avatar simonbaird avatar sonam1412 avatar staticf0x avatar stuartwdouglas avatar tisutisu avatar tkdchen avatar zregvart avatar

Watchers

 avatar

Recommend Projects

  • React photo React

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

  • Vue.js photo Vue.js

    ๐Ÿ–– 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. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

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 โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.