CMS Plone Chart for CDK8S

This chart provides a library to bootstrap a Plone deployment on a Kubernetes cluster using the CDK8S framework.

It provides

• Backend (as API with plone.volto or as Classic-UI)
• Frontend (Plone-Volto, a ReactJS based user interface)
• Varnish using kube-httpcache. It includes a way to invalidate varnish cluster (optional)

Typescript

To use this library, create a new CDK8S project (or use an existing one)

cdk8s init typescript-app

Then add the following dependency to package.json:

{
  "dependencies": {
    "@bluedynamics/cdk8s-plone": "*"
  }
}

Run npm install to install cdk8s-plone.

Python

Todo: Document in details how to install.

cdk8s init python-app

Python package name is cdk8s-plone.

Usage

With cdk8s-cli installed, create a new project:

cdk8s sythn

Add the following code to your main.ts:

...
import { Plone } from '@bluedynamics/cdk8s-plone';
...
    super(scope, id, props);

    // define resources here
    new Plone(this, 'Plone', {});
...

Run npm run build to generate the Kubernetes manifests. The manifests are stored in the dist directory.

For more have a look at the example project.

Prerequisites

For using cdk8s-plone, we assume you already have following tools installed:

• kubectl – A command-line tool for interacting with Kubernetes clusters. For deploying the Kubernetes manifest you will need a tool like this. Take a look at the Install Tools for kubectl.

• (optional) Helm – A Kubernetes package manager for managing Plone/Volto deployments. This tool is optional and only needed if you generate helm charts as output with cdk8s synth - instead of pure manifests. There are several ways to install it see the install section for Helm.

References

Kubernetes Documentation relevant for ressource management, readiness and liveness

PloneBaseOptions

Interface

• image(string):
  • The used Plone image
  • e.g. plone/plone-backend:6.1.0
• imagePullPolicy(string):
  • default IfNotPresent
• replicas(numbers)
• maxUnavailable(number|string)
• minAvailable(number|string)
• limitCpu(string)
• limitMemory(string)
• requestCpu(string)
• requestMemory(string)
• environment(kplus.Env)
• readinessEnabled(boolean)
• readinessInitialDelaySeconds(number)
• readinessIimeoutSeconds(number)
• readinessPeriodSeconds(number)
• readinessSuccessThreshold(number)
• readinessFailureThreshold(number)
• livenessEnabled(boolean)
  • should be true for volto
  • should be false for backend/classicui
• livenessInitialDelaySeconds(number)
• livenessIimeoutSeconds(number)
• livenessPeriodSeconds(number)
• livenessSuccessThreshold(number)
• livenessFailureThreshold(number)

PloneOptions

Interface

• version(string):
  • version of your project
• siteId(string):
  • default Plone
• variant(PloneVariant):
  • default PloneVariant.VOLTO
• backend (PloneBaseOptions):
  • default {}
  • needs image and enviroment
• frontend (PloneBaseOptions):
  • default {}
  • needs image if PloneVariant.VOLTO
• imagePullSecrets(string[])

PloneVariants

Enum

• VOLTO = 'volto'
• CLASSICUI = 'classicui'
  • no frontend options/image needed

Plone

class

builds the Plone Construct

• backendServiceName(string)
• frontendServiceName(string)
• variant(PloneVariant)
  • default Volto
• siteId(string)
  • default Plone

PloneHttpcacheOptions

Interface

• plone(Plone):
  • Plone chart
• varnishVcl{string}:
  • varnishfile
  • per default varnishVclFile should be used
• varnishVclFile(string):
  • File in config folder
• existingSecret(string)
• limitCpu(string)
• limitMemory(string)
• requestCpu(string)
• requestMemory(string)
• servicemonitor(string)
  • default false used for metrics

PloneHttpcache

class

uses helmchart kube-httpcache and builds the PloneHttpCache Construct

• scope(Construct)
• id(string)
• options(PloneHttpcacheOptions)

Development

Clone the repository and install the dependencies:

nvm use lts/*
corepack enable
npx projen

Then run the following command to run the test:

npx projen test

ToDo

• Option to enable Servicemonitor