Production infrastructure · Virtualization · SAN · Recovery

Enterprise VMware-to-Hyper-V migration

I moved production workloads from VMware ESXi to Windows Server 2025 Hyper-V while protecting the source SAN data, preserving VM identity, and validating the complete service before controlled cutover.

Context

The work crossed VMware, Windows Server 2025 Hyper-V, an enterprise SAN, Veeam, Active Directory, DNS, applications, and production networking. A separate conversion host provided an isolated conversion and validation point.

My role

I planned and carried out the conversion, storage-handling, isolated-testing, and validation steps across the migration path.

Result

The workloads moved with their identity, disk layout, networking, and application function preserved while the source remained a controlled recovery point.

Safe working sequence

How I reduced risk

  1. Discover
  2. Protect source
  3. Convert
  4. Isolate
  5. Cut over
  6. Verify

Conceptual workflow only. No live topology, identifiers, or operational data are shown.

The problem

Why the work needed more than one tool

A platform migration can look like a disk-conversion task when it is really a dependency and recovery problem. The virtual machines were tied to SAN presentation, network identity, domain services, applications, backup state, and the order in which source and target storage became visible.

What could go wrong

  • Writing to or initializing the wrong SAN LUN
  • Booting duplicate production identities on the live network
  • Losing MAC addresses, static network identity, disk order, or application dependencies
  • Treating a booted VM as proof that the service had recovered
  • Removing the source recovery path too early

What I did

The working sequence

  • Inventoried virtual machines, disks, datastores, LUNs, networks, backup state, and service dependencies
  • Designed temporary and permanent SAN layouts for conversion, restore staging, and final Hyper-V placement
  • Configured SAS initiators, host groups, LUN mappings, MPIO, and read-only source access
  • Used Windows SAN Policy OfflineAll and disabled automount before presenting storage
  • Configured the Hyper-V host as a Veeam-managed server and Direct SAN proxy
  • Restored multi-terabyte VMware virtual machines directly from SAN-backed datastores
  • Preserved VM identity, MAC addresses, disk layout, and static network configuration
  • Used isolated Hyper-V networks for boot and service testing before production cutover

How I validated it

Evidence before completion

  • Verified source LUN identity and read-only state before target initialization
  • Checked restore points, virtual disk count, VHDX placement, and checkpoint state
  • Tested guest boot, storage, networking, authentication, and application startup in isolation
  • For domain controllers, verified DNS, SYSVOL, NETLOGON, FSMO ownership, replication health, and static identity
  • Moved workloads into the final production state only after dependency checks passed

What this demonstrates

The skill behind the tools

This project demonstrates that I can move between virtualization, SAN storage, backup tooling, Windows identity, networking, and application validation without losing sight of production safety.

Production details are intentionally anonymized. I do not publish organization names, internal addresses, credentials, private topology, or sensitive recovery information.

Platforms and tools

VMware ESXi Windows Server 2025 Hyper-V Temporary conversion host enterprise SAN Veeam Backup & Replication SAS MPIO VMFS ReFS Active Directory

Continue through the project inventory