What is kubectl-mtv?

kubectl-mtv is a powerful command-line interface (CLI) plugin for Kubernetes that enables seamless migration of virtual machines from various virtualization platforms to KubeVirt-enabled Kubernetes clusters. MTV stands for “Migration Toolkit for Virtualization” and serves as the primary user interface for the Forklift project.

As a kubectl plugin, kubectl-mtv integrates directly with your existing Kubernetes workflow, providing a familiar command-line experience for managing complex VM migrations. It acts as a bridge between traditional virtualization infrastructure and modern container orchestration platforms, enabling organizations to modernize their workloads by migrating to Kubernetes-native virtualization.

For complete information about the underlying migration technology, see the official Forklift documentation.

Core Functionality and Supported Platforms

Supported Source Platforms

kubectl-mtv supports migration from the following virtualization platforms:

  • VMware vSphere - Full support for vCenter-managed VMware environments with advanced features like VDDK optimization (see VMware Prerequisites)
  • Red Hat Virtualization (oVirt/RHV) - Complete support for oVirt and Red Hat Virtualization platforms (see oVirt Prerequisites)
  • Red Hat OpenStack Platform - Support for OpenStack-based virtualization environments (see OpenStack Prerequisites)
  • Amazon EC2 - Support for migrating EC2 instances from AWS
  • KubeVirt/OpenShift Virtualization - Support for migrating VMs between KubeVirt clusters or within the same cluster (see KubeVirt Prerequisites)
  • OVA Files - Direct import and conversion of OVA (Open Virtualization Appliance) files (see OVA Prerequisites)

Target Platform

  • KubeVirt on Kubernetes/OpenShift - All migrations target KubeVirt-enabled Kubernetes clusters, supporting both upstream Kubernetes and Red Hat OpenShift platforms

Migration Types

The tool supports multiple migration strategies to accommodate different use cases:

  • Cold Migration - Traditional offline migration where the source VM is powered down during the process (see Forklift Cold Migration)
  • Warm Migration - Pre-copy migration that minimizes downtime by transferring data while the VM is running, followed by a brief cutover period (see Forklift Warm Migration)
  • Live Migration - Advanced migration with minimal downtime using KubeVirt’s live migration capabilities (available only for KubeVirt/OpenShift Virtualization sources) (see Forklift Live Migration)
  • Conversion Migration - Perform only guest conversion and VM creation when storage vendors provide pre-populated PVCs (VMware sources only) (see Chapter 3.6: Conversion Migration)

For detailed information about migration types, see:

Key Features

1. Advanced Query Language Integration

kubectl-mtv incorporates the powerful Tree Search Language (TSL), developed by Yaacov Zamir, which enables sophisticated filtering and searching of inventory resources. This is a kubectl-mtv-specific enhancement not available in the base Forklift web interface:

  • SQL-like syntax for intuitive resource queries
  • Complex filtering with support for logical operators (AND, OR, NOT)
  • Pattern matching with LIKE and ILIKE operators
  • Comparison operators for numeric and date-based filtering
  • Built-in functions such as LEN, ANY, ALL, and SUM

Example query syntax:

1

kubectl mtv get inventory vms vsphere-01 -q "where memoryMB > 4096 and powerState = 'poweredOn'"

2. Flexible Mapping System

The tool provides multiple approaches to resource mapping:

  • Explicit Mappings - Pre-defined network and storage mappings for consistent, reusable configurations
  • Inline Mapping Pairs - Direct specification of source-to-target mappings during plan creation
  • Automatic Default Mappings - Intelligent mapping based on available target resources
  • Enhanced Storage Options - Support for volume modes, access modes, and advanced storage array offloading for up to 10x faster migrations (see Chapter 9.5: Storage Array Offloading)

3. VDDK (Virtual Disk Development Kit) Support

For VMware environments, kubectl-mtv provides optimized disk transfer capabilities:

  • Custom VDDK Image Creation - Build optimized disk transfer containers
  • Performance Optimization - Significantly faster disk transfers compared to standard methods
  • Advanced Configuration - Support for buffer sizing and AIO optimization
  • Seamless Integration - Automatic VDDK usage when configured

4. Real-time Monitoring and Management

Comprehensive lifecycle management for migrations:

  • Live Progress Tracking - Real-time monitoring with --watch flag
  • Detailed Status Reporting - Complete visibility into migration phases and steps
  • Migration Control - Start, pause, resume, cancel, and cutover operations
  • Archive Management - Archive completed migrations for historical tracking

5. Kubernetes-native Resource Management (KARL)

Integration with the Kubernetes Affinity Rule Language (KARL) for advanced pod placement. This is a kubectl-mtv-specific enhancement that simplifies complex Kubernetes affinity expressions:

  • Declarative Affinity Rules - Express complex placement requirements using natural language syntax
  • Topology-aware Placement - Support for node, zone, region, and rack-level affinity
  • Multiple Rule Types - REQUIRE (hard), PREFER (soft), AVOID (hard anti-affinity), REPEL (soft anti-affinity)
  • Target VM and Convertor Pod Optimization - Separate affinity rules for runtime VMs and migration workloads

Example KARL syntax:

1
2

--target-affinity "REQUIRE pods with app=database on node"
--convertor-affinity "PREFER pods with storage=high-performance on zone"

6. Migration Hooks and Automation

Support for custom automation through migration hooks:

  • Pre-migration Hooks - Execute custom logic before migration begins
  • Post-migration Hooks - Perform cleanup or configuration after migration completes
  • Ansible Playbook Integration - Run Ansible playbooks as part of the migration process
  • Shell Script Support - Execute custom shell scripts with access to migration context
  • Context Access - Hooks receive detailed migration context including plan and workload information

7. Model Context Protocol (MCP) Integration

Advanced AI assistant integration for enhanced user experience. This is a kubectl-mtv-specific feature not available in standard Forklift:

  • AI Assistant Access - Provide Claude, Cursor IDE, and other MCP-compatible tools access to migration resources
  • Multiple Server Modes - Support for both stdio and SSE (Server-Sent Events) communication
  • Natural Language Queries - Enable AI assistants to interpret and execute migration commands
  • Intelligent Assistance - AI-powered recommendations and troubleshooting support

Relationship with Forklift/Migration Toolkit for Virtualization (MTV)

kubectl-mtv serves as the primary command-line interface for the Forklift project, which is Red Hat’s upstream migration toolkit. The relationship between these components is as follows:

Forklift Controller

  • Backend Engine - Handles the actual migration orchestration and execution (see Forklift Architecture)
  • Custom Resource Management - Manages Kubernetes custom resources for providers, plans, and mappings (see Forklift Custom Resources)
  • Integration Layer - Interfaces with source virtualization platforms and target KubeVirt infrastructure

kubectl-mtv CLI

  • User Interface - Provides intuitive command-line access to Forklift functionality
  • Resource Management - Simplifies creation, modification, and monitoring of migration resources
  • Workflow Orchestration - Streamlines complex migration workflows into simple command sequences
  • Advanced Features - Adds sophisticated query capabilities (TSL), affinity management (KARL), hooks, and AI integration

Migration Toolkit for Virtualization (MTV)

  • Product Integration - MTV is Red Hat’s supported product offering that includes Forklift
  • Enterprise Features - Additional enterprise-grade features, support, and integration
  • OpenShift Integration - Deep integration with Red Hat OpenShift Container Platform

Architecture Flow

  1. Resource Definition - Users define providers, mappings, and plans using kubectl-mtv
  2. API Translation - Commands are translated to Kubernetes API calls creating Forklift custom resources
  3. Controller Processing - Forklift controllers process these resources and orchestrate migrations
  4. Status Reporting - Migration status flows back through the API to kubectl-mtv for user visibility

This architecture ensures that kubectl-mtv provides a powerful, user-friendly interface while leveraging the robust, battle-tested migration engine provided by Forklift. The separation of concerns allows for specialized optimization in both the user experience and the migration execution layers.

Getting Started

With this foundational understanding of kubectl-mtv, you’re ready to proceed to the next chapter covering installation and prerequisites. The tool’s comprehensive feature set enables everything from simple single-VM migrations to complex enterprise-scale migration projects with hundreds of virtual machines.

The following chapters will guide you through:

  • Detailed installation procedures
  • Provider setup and configuration
  • Advanced querying and filtering techniques
  • Mapping creation and management
  • Migration plan development and execution
  • Optimization strategies for performance and placement
  • Troubleshooting and best practices

Next: Chapter 2: Installation and Prerequisites