KUDO CLI

This section describes how to install the Kubernetes Universal Declarative Operator (KUDO) CLI and initialize the KUDO cluster component.

What Is KUDO

KUDO provides a declarative approach to building and installing production-grade Kubernetes operators. To quote the official documentation: “Operators are software extensions to Kubernetes that make use of custom resources to manage applications and their components.” While Kubernetes already comes with a lot of built-in automation to run simple workloads, complex scenarios often need a human operator. This is where KUDO is designed to help.

Key Features

Key features of KUDO are:

Repository of managed packages
Custom repositories for own packages
Mechanisms for upgrading packages
Templating support focused on on Day-2 operations. KUDO can configure an entire Operator’s lifecycle using a declarative spec. This allows for further interactions with installed applications like triggering a backup/restore.

Before you begin

You need certain software configurations and settings before you start this procedure. This procedure requires the following items and configurations:

A running Kubernetes cluster

Install Kudo using krew

Using krew, the package manager for kubectl plugins, is the easiest way to install KUDO CLI. Install KUDO CLI by entering the following command:

kubectl krew install kudo

Check if KUDO CLI installed properly. Enter the following command:

kubectl kudo help

Install KUDO on MacOS

Use brew to install KUDO CLI on MacOS. Enter the following command:

brew tap kudobuilder/tap
brew install kudo-cli

Install KUDO on Linux

Use latest release page to download the CLI binaries for your platform. Use this script to download the correct binary and add it to your path:

VERSION=x.y.z # look up the current stable release at https://github.com/kudobuilder/kudo/releases/latest
OS=$(uname | tr '[:upper:]' '[:lower:]')
ARCH=$(uname -m)
wget -O kubectl-kudo https://github.com/kudobuilder/kudo/releases/download/v${VERSION}/kubectl-kudo_${VERSION}_${OS}_${ARCH}
chmod +x kubectl-kudo
# add to your path
sudo mv kubectl-kudo /usr/local/bin/kubectl-kudo

Initialize KUDO

Unlike Helm, KUDO has an installed component, called KUDO Manager, inside the cluster. Install KUDO Manager. Enter the following commands:

kubectl kudo init

This command installs all the cluster components required for KUDO to work. The init command aborts if it detects an existing KUDO installation, preventing accidental upgrades. For more information on how to customize the KUDO initialization, see the init command. To upgrade an existing KUDO installation, see kudo upgrade.

Install a KUDO Operator

There are several ways to install a KUDO operator. KUDO comes with a preconfigured official operators repository.

As an example, install an instance of Zookeeper operator named zk by entering the following command:

kubectl kudo install zookeeper --instance zk

Check the status of the instance. Enter the following command:

$ kubectl kudo plan status --instance zk
Plan(s) for "zk" in namespace "default":
.
└── zk (Operator-Version: "zookeeper-0.3.1" Active-Plan: "deploy")
    ├── Plan deploy (serial strategy) [COMPLETE], last updated 2020-10-07 14:12:55
    │   ├── Phase zookeeper (parallel strategy) [COMPLETE]
    │   │   └── Step deploy [COMPLETE]
    │   └── Phase validation (serial strategy) [COMPLETE]
    │       ├── Step validation [COMPLETE]
    │       └── Step cleanup [COMPLETE]
    ...

KUDO operators can have multiple plans, defined by the operator developer, for the user to execute. Every operator must contain at least a deploy plan. This is the default plan to deploy an application to the cluster. In the above output, Zookeepers deploy plan consists of two phases. Each of these phases have several steps which have completed by now.

For information on KUDO, refer to the following:

KUDO CLI Usage