Skip to content

Deployment automation for applications on kubernetes.

License

Notifications You must be signed in to change notification settings

gapfish/deployer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Deployer, deployment automation for applications on kubernetes

Architecture

Configuration

You can find a sample config file here config.sample.yml. Sample configuration:

repositories:
  - name: cool_app
    github: schasse/cool_app
    kube_resource: kubernetes

If you use private github repositories, a private dockerhub registry and want to secure your Deployer from unrestricted access, it is necessary to configure secrets via environment variables. You can configure those three secrets:

# .env
export DEPLOYER_GITHUB_TOKEN="df16f9887f5a45980ca37af7c9a1d1100e477a0f"
export DEPLOYER_DOCKER_REGISTRY_CREDS="dockeraccount:du9oothiethah3Faelie"
export DEPLOYER_AUTH_TOKEN="6ab24b5c7e3cb7b19de93565f52208d10f58cc39"

You can use several different solutions to configure Deployer on Kubernetes. In the following there are the two recommended solutions.

Configuration via Docker image

The first solution is to build an own docker image and add the config.yml to the image. See the following Dockerfile as an example.

FROM gapfish/deployer:latest
COPY .env .
COPY config.yml .

Configuration via Secret and ConfigMap

Another option is to mount the configuration via a ConfigMap. Create a Kubernetes ConfigMap like the following one:

# kubernetes/configmap.yml
apiVersion: v1
kind: ConfigMap
metadata:
  name: deployer-config
data:
  config.yml: |
    repositories:
    - name: cool_app
      github: schasse/cool_app
      kube_resource: kubernetes

You'll also need to adjust the Deployment to mount the config.yml. See the following snippet therefor:

# kubernetes/deploy.yml
...
  volumeMounts:
  - name: env
    mountPath: /deployer/.env
    subPath: .env
  - name: config
    mountPath: /deployer/config.yml
    subPath: config.yml
...
volumes:
- name: config
  configMap:
  name: deployer-config
  items:
  - key: config.yml
    path: ./config.yml
- name: env
  secret:
    secretName: deployer-env

Then create all the resources in your Kubernetes cluster.

kubectl create secret generic deployer-env --from-file=.env=.env
kubectl create -f kubernetes/*

Deployment on Kubernetes

You can easily deploy Deployer on Kubernetes and after the initial setup the Deployer can even deploy itself.

Have a look at the example Kubernetes resources to deploy deployer. Therefor see kubernetes/ folder.

depctl CLI

Installation

Docker is a prerequisite for depctl. If you have docker installed, simply dowload depctl via:

curl https://raw.githubusercontent.com/gapfish/deployer/master/depctl/bin/depctl | sudo tee /usr/local/bin/depctl

To update depctl run docker pull gapfish/depctl.

Configuration

depctl must be configured by a configuration file located at $HOME/.deployer/deployer_env. See the following configuration as an example.

# ~/.deployer/deployer_env
export DEPLOYER_AUTH_TOKEN="xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
export DEPLOYER_URL="https://deployer.my-domain.com"

Usage

Directly control the Deployer via depctl Api CLI.

~ depctl --help

Commands:
  build                ALPHA FEATURE: Builds a repository locally and then
                       pushes to registry.
  canary [REPOSITORY]  Deploy a canary for a repository.
  deploy [REPOSITORY]  Deploy a repository.
  ls                   List available repositories.
  show [REPOSITORY]    Show the repository.
  tags [REPOSITORY]    List tags for a repository.
  ------------------------------------------------------------------------
  help [(tags|deploy|canary)] Prints help for the command.
  version                     Prints the depctl and ruby version.

About

Deployment automation for applications on kubernetes.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages