Skip to content

Latest commit

 

History

History
128 lines (87 loc) · 8.14 KB

CONTRIBUTING.md

File metadata and controls

128 lines (87 loc) · 8.14 KB

Contributing Guidelines

First: if you're unsure or afraid of anything, just ask or submit the issue or pull request anyway. You won't be yelled at for giving your best effort. The worst that can happen is that you'll be politely asked to change something. We appreciate any sort of contributions, and don't want a wall of rules to get in the way of that.

However, for those individuals who want a bit more guidance on the best way to contribute to the project, read on. This document will cover what we're looking for. By addressing all the points we're looking for, it raises the chances we can quickly merge or address your contributions.

Sign the CLA

Kubernetes projects require that you sign a Contributor License Agreement (CLA) before we can accept your pull requests.

Please see https://git.k8s.io/community/CLA.md for more info.

Contributing Steps

  1. Submit an issue describing your proposed change to the repo in question.
  2. The repo owners will respond to your issue promptly.
  3. If your proposed change is accepted, and you haven't already done so, sign a Contributor License Agreement (see details above).
  4. Fork the desired repo, develop and test your code changes.
  5. Submit a pull request.

How to build Kudo locally

Pre-requisites

  • Git
  • Go 1.13 or later
  • Kubebuilder version 2 or later - note that it is only needed for the kube-apiserver and etcd binaries, so no need to install its dependencies (such as kustomize).
  • A Kubernetes Cluster running version 1.13 or later (e.g., kind or Minikube)
  • kubectl

Build Instructions

  • Get the KUDO repo: git clone https://github.com/kudobuilder/kudo.git
  • cd kudo
  • Export GOPATH (this is necessary because of an issue in codegenerator)
  • make all to build manager as well as CLI
  • [optionally] make docker-build to build the manager Docker images

When updating the structs under APIs, or any other code generated item, use make generate to generate the new DeepCopy structs.

Running manager locally

The most convenient way to test new controller code is to run the manager locally. It will use kubernetes cluster defined via your local kubeconfig to talk to API server and resolve CRDs. You can run manager locally via make run.

Make sure your local cluster has up to date CRDs. You can deploy new CRDs with make deploy. Beware that make deploy also deploys manager into your cluster (kubectl get deployments -n kudo-system) and it will be the latest stable manager, not the one from your current git. If you plan to run your own manager, just delete the one in your cluster via kubectl delete deployment kudo-controller-manager -n kudo-system

Testing new CLI

You can build CLI locally via make cli. After running that command, CLI will be available in bin/kubectl-kudo and you can invoke the command for example like this bin/kubectl-kudo init (no need to install it as kubectl plugin).

Running new manager inside cluster

For some situations, it might make sense to test your manager inside a real cluster running in a pod (not just running the binary locally). To do that you need:

  • build a docker image with the manager locally DOCKER_IMG=nameofyourimage make docker-build
  • push the image to a remote repository DOCKER_IMG=nameofyourimage make docker-push
  • run kubectl kudo init --kudo-image nameofyourimage:tag

Testing

See the contributor's testing guide.

Community, Discussion, and Support

Learn how to engage with the KUDO community on the community page.

You can reach the maintainers of this project at:

Code culture

This is a set of practices we try to live by when developing KUDO. These are just defaults (soft rules). Deviations from them are possible, but have to be justified.

General guidelines

  • Master is always releasable (green CI)
  • All feature/bug-fixing work should have an open issue with a description, unless it's something very simple
  • Every user-facing feature that is NOT behind a feature gate should have integration or an e2e test

Pull requests

  • One core-team member has to approve the PR to be able to merge (all people listed in .github/CODEOWNERS file)

  • One approval is enough to merge. However, if there are requests for change they have to be resolved prior to the merge

  • Since KUDO is developed in multiple timezones, try to keep the PR open for everyone to be able to see it (~24h, keep in mind public holidays)

  • We prefer squash commits so that all changes from a branch are committed to master as a single commit

  • Before you merge, make sure your commit title and description are meaningful. Github by default will list all the individual PR commits when squashing which are rarely insightful. We aim for a clean and meaningful commit history.

  • For a piece of work that takes >3-5 days, pair with somebody

  • When you pair with somebody, don't forget to appreciate their work using co-authorship

  • Open a PR as soon as possible to give everybody a chance to review it

  • For PRs that tackle a bigger feature/refactoring schedule a walk-through with the team. PR reviews are a lot more meaningful if reviewers understand your code mental model.

As a code owner (core team member)

  • Schedule a portion of your day to review PRs to appreciate work of others

Is My Thing an Enhancement?

We are trying to figure out the exact shape of an enhancement. Until then here are a few rough heuristics.

An enhancement is anything that:

  • a blog post would be written about after its release (ex. minikube, StatefulSets, rkt container runtime)
  • requires multiple parties/owners participating to complete
  • will be graduating from one stage to another (ex. alpha to beta, beta to GA)
  • needs significant effort or changes KUDO in a significant way (ex. something that would take 10 person-weeks to implement, introduce or redesign a system component)
  • impacts the UX or operation of KUDO substantially such that engineers using KUDO will need retraining
  • users will notice and come to rely on

It is unlikely an enhancement if it is:

  • fixing a flaky test
  • refactoring code
  • performance improvements, which are only visible to users as faster API operations, or faster control loops
  • adding error messages or events

If you are not sure, ask someone in the #kudo channel on Slack or ping someone listed in CODEOWNERS.

When to Create a New Enhancement Issue

Create an issue in this repo once you:

  • have circulated your idea to see if there is interest
    • through Community Meetings, KUDO meetings, KUDO mailing lists, or an issue in github.com/kudobuilder/kudo
  • (optionally) have done a prototype in your own fork
  • have identified people who agree to work on the enhancement
    • many enhancements will take several releases to progress through Alpha, Beta, and Stable stages
    • you and your team should be prepared to work on the approx. 9mo - 1 year that it takes to progress to Stable status
  • are ready to be the project-manager for the enhancement

Code of Conduct

Participation in the Kubernetes community is governed by the Kubernetes Code of Conduct.