Jump to content

Connect Leeroopedia MCP: Equip your AI agents to search best practices, build plans, verify code, diagnose failures, and look up hyperparameter defaults.

Environment:Kubeflow Kubeflow Kubectl Kustomize CLI Environment

From Leeroopedia
Knowledge Sources
Domains Infrastructure, CLI_Tools, Platform_Deployment
Last Updated 2026-02-13 00:00 GMT

Overview

Local workstation environment with kubectl (>= 1.25) and kustomize CLI tools for deploying and managing Kubeflow.

Description

This environment defines the local CLI tools required on the operator's workstation for deploying Kubeflow to a Kubernetes cluster. The kubectl CLI must be version 1.25 or higher to support the API resources used by Kubeflow. The kustomize standalone binary is required to render Kubeflow manifests using the kustomize build command. Note that the kustomize built into kubectl (kubectl apply -k) may lag behind the standalone version and is not recommended for Kubeflow deployments.

Usage

Use this environment for all Platform Deployment workflows and for ongoing cluster management tasks (health checks, scaling, debugging). Both tools must be on the system PATH before running any Kubeflow deployment commands.

System Requirements

Category Requirement Notes
OS Linux, macOS, or Windows (WSL2) Linux is the primary deployment platform
kubectl >= 1.25 Must match within one minor version of the target cluster
kustomize >= 5.0.0 (standalone) Do not use kubectl apply -k; use standalone kustomize
Disk Minimal (< 500MB) Just the CLI binaries and kubeconfig
Network Access to Kubernetes API server KUBECONFIG must point to a reachable cluster

Dependencies

System Packages

  • kubectl >= 1.25
  • kustomize >= 5.0.0 (standalone binary)
  • curl or wget (for downloading binaries)
  • git (for cloning kubeflow/manifests repository)

Credentials

The following environment variables may be set:

  • KUBECONFIG: Path to kubeconfig file with cluster access credentials. Defaults to ~/.kube/config.

Quick Install

# Install kubectl (Linux amd64)
curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl"
chmod +x kubectl
sudo mv kubectl /usr/local/bin/

# Install kustomize (standalone)
curl -s "https://raw.githubusercontent.com/kubernetes-sigs/kustomize/master/hack/install_kustomize.sh" | bash
sudo mv kustomize /usr/local/bin/

# Verify
kubectl version --client
kustomize version

Code Evidence

Installation instructions referenced from README.md:L46-47:

The Kubeflow AI reference platform can be installed via Packaged Distributions
or Kubeflow Manifests.

The deployment pattern kustomize build ... | kubectl apply -f - is the standard Kubeflow Manifests installation method, documented in the kubeflow/manifests repository and referenced by the README.

Version check implementation from the Prerequisites Validation step:

kubectl version --client --output=yaml
kubectl version --output=yaml
kustomize version

Common Errors

Error Message Cause Solution
command not found: kubectl kubectl not installed or not on PATH Install kubectl and ensure it is in /usr/local/bin/ or on PATH
command not found: kustomize kustomize not installed Install standalone kustomize binary (do not rely on kubectl built-in)
error: unable to recognize STDIN: no matches for kind kustomize version too old for the manifests Upgrade to kustomize >= 5.0.0
error: the server doesn't have a resource type kubectl client version mismatch with server Upgrade kubectl to within one minor version of the cluster

Compatibility Notes

  • macOS: Use brew install kubectl kustomize for Homebrew-based installation.
  • Windows: Use WSL2 with Ubuntu. Native Windows kubectl works but kustomize may have path issues.
  • kubectl built-in kustomize: The kubectl apply -k command uses an embedded kustomize that may be older than the standalone version. Kubeflow manifests often require features only available in the standalone kustomize binary.

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment