kubic-ci 0.4.6

# ci3

Continuous deployment for kubernetes (k8s) and gitlab.

# Table of Contents 1. [blog](#blog) 2. [use](#use) 3. [develop](#develop)

## blog

More articles on [Medium..](https://medium.com/kubic-ci)

## use

Prerequisites: ci3 is written and tested in python3 (linux or mac osx).

As a user you would probably do something like:

pip install kubic-ci

The following should work w/o errors:

kubic –help

To start a new project locally you would proceed with:

kubic init

A typical deploy cycle is:

kubic build && kubic push && kubic deploy -d hello

where hello is the name of your service and project.

### docker image

One can find a simple docker image inside the docker folder, which is suitable to be used by gitlab-runner, so that one has already ci3 preinstalled and can deploy to GKE.

The following will pull latest image from dockerhub and run it in your local docker

docker run -it kubic3/ci3:latest

Like this kubic-ci can be integrated into CI/CD cycle provided by [gitlab](https://docs.gitlab.com/ee/ci/yaml/).

## develop

### release

Before making a new release:

1. Make sure tests are passing (both locally and remotely on gitlab CI).

2. Increment version according to [semantic versioning](https://semver.org/).

In addition to bumping a new version in ci3/version.py, issuing a new release includes few more publishing (post-release) activities:

1. Publish new python package python setup.py sdist upload

2. Build ad upload a new docker image (to avoid confusions, both docker image and python package share the same version). Check how it has been picked up on [dockerhub](https://hub.docker.com/r/kubic3/ci3/) and [microbadger](https://microbadger.com/images/kubic3/ci3).

3. Say something on socials about it ;)