DKube Installation & Cluster Management

Contents:

• Getting Started
  • Overview
    • DKube Configuration
    • Installation Configuration
      • DKube and Kubernetes
  • Prerequisites
    • Supported Platforms
    • Node Requirements
      • Installation Node Requirements
      • DKube Cluster Node Requirements
    • Cluster Access
  • Getting the DKube Files
  • Platform-Specific Installation Instructions
• DKube-Specific Steps for an Existing for EKS Cluster
  • EKS Cluster Installation Files
    • Running the EKS Preinstallation Script
  • DKube Installation
• DKube-Specific Steps for an Existing Rancher Cluster
  • Setting up the Rancher Cluster
    • Updating the cluster.yml File
    • Execute the Rancher Server Run Command
    • Copying the Kubeconfig file to the DKube Cluster Installation Node
  • Preparing the Rancher Cluster
    • Editing the k8s ini File
      • User
      • Changing Nodes or GPUs on Cluster After DKube Installation
    • Cluster Access from the Installation Node
    • Platform-Specific Cluster Access
    • Final Access Verification
  • Node Setup
  • Installing DKube
• DKube-Specific Steps for an Existing VMWare Tanzu Cluster
  • Getting the kubeconfig file
  • Preparing the Cluster
    • Editing the k8s ini File
      • User
      • Changing Nodes or GPUs on Cluster After DKube Installation
    • Cluster Access from the Installation Node
    • Final Access Verification
  • Node Setup
  • Installing DKube
• Installing DKube
  • Getting the Helm Files
  • Express Installation
    • Installation Dashboard
    • Accessing DKube
    • Initial Login
• Advanced Installation Options
  • Basic Configuration
    • Resilient Operation
      • DKube Resiliency
      • Resiliency Examples
    • Username and Password
  • Storage Options
  • Load Balancer Options
  • CI/CD
  • Node Affinity
    • Job Type-Related Affinity
    • GPU-Related Affinity
    • Workload-Related Affinity
    • Node Affinity Usage
    • Assigning a Label
    • Assigning a Taint
  • Model Monitor
  • External Database
  • Helm-Based DKube Installation
    • Installation Status
  • Installation Dashboard
    • Dashboard Status
  • Accessing DKube
    • Initial Login
• Managing DKube
  • Upgrading DKube
  • Uninstalling DKube
  • Reinstalling DKube
  • Backup and Restore
    • Backup Exclusions
    • Editing the backup.ini File
    • Backup
    • List of Snapshots
    • Delete Backup Snapshot
    • Restore
      • Restore Failure
  • DKube Migration
    • Editing the Migration.ini File
      • Getting the Access Token
    • Executing the Migration
  • Stop and Start DKube
    • Stopping DKube
    • Starting DKube
  • Restarting DKube After Cluster Restart
    • Docker Status
• Managing the Kubernetes Cluster
  • Adding A100 GPUs to the Cluster
    • Enable MIG Devices
    • List GPU Instance Profiles
    • Create MIG Devices
    • Delete MIG Instances
  • Changing Nodes or GPUs
• Software Package Help
  • Docker Installation on Ubuntu
  • Docker Installation on CentOS
  • kubectl Installation
  • Removing Nouveau Driver
• Troubleshooting
  • During Installation, Access to Cluster System is Denied
    • Symptom
    • Cause/Action
  • Problems Accessing DKube url
    • Symptom
    • Cause/Action
  • GPUs are Not Available on Master Node
    • Symptom
    • Cause/Action
  • Users do Not Show up at On Boarding Screen
    • Symptom
    • Cause/Action