Skip to content
This repository has been archived by the owner on Oct 15, 2018. It is now read-only.

Helm Chart Publisher is a HTTP API for publishing Helm Charts in a easy way

Notifications You must be signed in to change notification settings

luizbafilho/helm-chart-publisher

Repository files navigation

Helm Chart Publisher

Helm Chart Publisher aims to help you build a nice CI/CD pipeline. It seats in front of a object storage service (such as AWS S3, OpenStack Swift) or a filesystem, sends your charts to it and also updates the index.

After receiving a PUT request with a repository and the chart, the publisher will upload the chart file to your storage, update the index and upload it too. Currently, it supports Amazon S3, OpenStack Swift, and Google Cloud Storage. Filesystem storage is planned.

Configuration

The configuration is based on a YAML file. In order to publish your charts, you have to configure a storage and one or more repos (Helm repositories).

The Helm repository isolation can be done via bucket or a directory. The publisher will create an index.yaml for each repository you configure.

Each repo requires name and bucket. You can also specify a directory, if you do so, the charts are going to be stored in bucket under the specified path.

Finally, you can also specify a url in the case you are using a proxy, CDN, etc. to direct traffic to your bucket. This also allows you the ability to add basic authentication in front of your repository in the form of https://[username]:[password]@[domain]. You can also specify subdirectories in the path if you so choose. The final output with a specified url will be [url]/[chart|index].

These are the configuration options for the helm publisher.

repos:
  - name: stable
    bucket: charts-bucket
  - name: incubator
    bucket: charts-bucket-incubator
  - name: test
    bucket: test-bucket
    directory: test
    url: https://[username]:[password]@mycharts.com/test

storage:
  gcs: {} # uses GCloud Application Default Credentials
  s3:
    bucket: charts-bucket-incubator
    region: us-west-2
  swift:
    username: SWIFT_USERNAME
    password: SWIFT_USERNAME
    tenant: some_tenant
    authUrl: https://some-auth-url:5000/v2.0
    endpointType: admin
    container: "kube-charts"
    insecureSkipVerify: false

Running

You can either:

  • Run as a binary
  • Run as a Docker container

Binary

Get the latest helm-chart-publisher for your platform on the releases page

curl -o /usr/local/bin/helm-chart-publisher -sSL https://github.com/luizbafilho/helm-chart-publisher/releases/download/<version>/helm-chart-publisher_<os>-<arch>
chmod +x /usr/local/bin/helm-chart-publisher

To run helm-chart-publisher you just have to execute the binary passing providing the configuration file.

$ PORT=8080 helm-chart-publisher --config /etc/helm-publisher/config.yaml

Docker Container

To run the helm-chart-publisher from within a container run make docker. This will create the helm-chart-publisher image.

Start the container using a command in the format:

docker run -p 8080:8080 -v /Users/$(whoami)/config:/etc/helm-chart-publisher/ -v /Users/$(whoami)/.aws:/root/.aws helm-chart-publisher
  • The publisher runs on 8080 within the container and the ports must be exposed
  • The config YAML must be mounted into the /etc/helm-chart-publisher directory
  • .aws configuration should also be mounted in to prevent restarting container when credentials expire

Usage

You can publish a chart calling a simple curl command.

$ curl -i -X PUT -F repo=stable -F chart=@$HOME/charts/stable/mariadb-0.5.9.tgz http://localhost:8080/charts

This command will upload the chart file to an Amazon S3 bucket, updates the current index.yaml and upload it too.

The indexes are available via publisher under the /:repo/index.yaml path. For example to access the stable index.

$ curl -i http://localhost:8080/stable/index.yaml

But you can still access the index.yaml going directly to the storage. In this case:

$ curl -i https://s3-us-west-2.amazonaws.com/charts-bucket/index.yaml

apiVersion: v1
entries:
  mariadb:
  - created: 2017-03-07T17:36:04.782813678-03:00
    description: Fast, reliable, scalable, and easy to use open-source relational
      database system. MariaDB Server is intended for mission-critical, heavy-load
      production systems as well as for embedding into mass-deployed software.
    digest: d68c2852d7ac3e431cc65278d3ab74946b28479319a5707bc4405adf1dcd1393
    engine: gotpl
    home: https://mariadb.org
    icon: https://bitnami.com/assets/stacks/mariadb/img/mariadb-stack-220x234.png
    keywords:
    - mariadb
    - mysql
    - database
    - sql
    maintainers:
    - email: [email protected]
      name: Bitnami
    name: mariadb
    sources:
    - https://github.com/bitnami/bitnami-docker-mariadb
    urls:
    - https://s3-us-west-2.amazonaws.com/charts-bucket/mariadb-0.5.9.tgz
    version: 0.5.9
generated: 2017-03-07T17:34:47.965508312-03:00

Roadmap

  • Storages
    • Openstack Swift
    • Google Cloud Storage
    • Filesystem
  • Tests
    • api
    • publisher
    • storages

Notes

This project is at a very early stage, suggestions are, as always, very welcome in the form of PR's. If you feel the documentation is not clear or you have any questions, please open an issue for that.