@npm-wharf/hikaru

A continuous deployment service and integration helper for kubernetes

Usage no npm install needed!

<script type="module">
  import npmWharfHikaru from 'https://cdn.skypack.dev/@npm-wharf/hikaru';
</script>

README

hikaru

A deployment tool for kubernetes. 100 internets if you get the reference. 1000 internets if you get why.

Build Status Coverage Status

What It Does - Deployments and Continuous Delivery

hikaru was built specifically to make it easy to deploy and maintain kubernetes clusters.

  • Initial deployment
  • Re-deployment/Updates

Features

  • Deploy mcgonagall specs from local directory
  • Supports tokenized specifications

Credential Limitations

hikaru uses kubectl under the hood and requires that you have a context configured to target each cluster you wish to deploy to.

Complimentary Tooling

hikaru requires mcgonagall style cluster specifications.

hikaru also requires that kubectl is installed and available on the PATH.

Modes

hikaru provides 2 possible models for interaction:

  • a CLI for interacting with Kubernetes clusters directly
  • custom use-cases via module

CLI

Full argument names are shown in the command examples. Shorthand arguments are available, see the interactive CLI help to get a list.

Installation

npm i @npm-wharf/hikaru -g

Tokens

If tokens are present in a specification and not provided via a tokenFile, hikaru will error and exit, informing you what tokens must be specified. It will refuse to run until all tokens have been defined.

Deploying Clusters (deploy)

Hikaru deploys specs through kubectl apply and so changes will be intelligently merged by the cluster.

Available options

  • --context (-c): the name of the kubectl context to deploy to. required
  • --tokenFile (-f): path to a file to read for tokens consumed in the spec.
  • --scale (-s): scale factor label to apply, if the given label is not used in your spec this will be ignored.
  • --verbose: output additional debug information.

Example usage:

hikaru deploy {spec} \
  --tokenFile {path to json, yaml or toml token file} \
  --context {name of kubectl context} \
  --scale {scaleLabel} \
  --verbose

Running Jobs

The run command will delete and apply the job resource again, forcing a new instance to run.

Available options

  • --job (-j): the name of the job resource to run in the format name.namespace required
  • --context (-c): the name of the kubectl context to deploy to. required
  • --tokenFile (-f): path to a file to read for tokens consumed in the spec.
  • --scale (-s): scale factor label to apply, if the given label is not used in your spec this will be ignored.
  • --verbose: output additional debug information.

Example usage:

hikaru run {spec} \
  --job {name.namespace, e.g. route53bot.infra} \
  --tokenFile {path to json, yaml or toml token file} \
  --context {name of kubectl context} \
  --scale {scaleLabel} \
  --verbose

Library

Deploying Cluster

const hikaru = require('hikaru')

hikaru.deploy({
  spec: '/path/to/spec',
  version: '1.13', // kubernetes api version
  data: {
    // tokens to apply to the spec are defined here
  },
  scale: 'medium' // scaleFactor
}) // returns a Promise

Running a Job

const hikaru = require('hikaru')

hikaru.run({
  spec: '/path/to/spec',
  job: 'name.namespace', // the job to run
  version: '1.13', // kubernetes api version
  data: {
    // tokens to apply to the spec are defined here
  },
  scale: 'medium' // scaleFactor
}) // returns a Promise