Production infrastructure · SAN · Source protection

Handling unknown SAN storage without touching live data

I developed a cautious workflow for discovering and using SAN storage from Windows without accidentally initializing, mounting, or writing to live VMware VMFS volumes.

Context

A Windows Hyper-V and Veeam workflow needed access to SAN-backed source and target storage. Some disks were unfamiliar from the Windows side, and a mistaken initialization could damage production data.

My role

I established the discovery controls, identified disks by authoritative attributes, protected source LUNs, prepared only confirmed targets, and validated the resulting restore path.

Result

Source VMFS volumes remained offline and protected while confirmed target LUNs were prepared as dedicated ReFS storage for Hyper-V and Veeam workloads.

The problem

Why the work needed more than one tool

Windows can discover SAN disks before an operator has proven which LUN is source, target, or unrelated production storage. Convenience features such as automount turn uncertainty into write risk.

What could go wrong

  • Initializing the wrong disk
  • Allowing Windows to mount or write to a source VMFS LUN
  • Relying on disk number or capacity alone
  • Presenting source and target storage without a clear map
  • Formatting before serial and SAN-side identity agree

What I did

The working sequence

  • Set Windows SAN Policy to OfflineAll before storage discovery
  • Disabled automount
  • Collected serial numbers and storage attributes before any initialization
  • Matched Windows-visible disks to SAN-side LUN and host mappings
  • Kept source VMFS LUNs offline and read-only
  • Initialized only confirmed target LUNs
  • Created dedicated ReFS volumes with 64 KB allocation units
  • Separated virtual machines, virtual disks, and restore staging in the target layout

How I validated it

Evidence before completion

  • Verified serial numbers and mappings from both Windows and SAN perspectives
  • Confirmed source disks remained offline and without Windows filesystems
  • Checked that only intended targets were online and writable
  • Validated ReFS allocation and folder layout
  • Completed isolated restore tests before using the storage for controlled production cutover

What this demonstrates

The skill behind the tools

This work demonstrates disciplined storage handling under uncertainty: stop automatic behavior, prove identity, protect the source, and only then prepare the target.

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

Platforms and tools

enterprise SAN Windows Server SAN Policy OfflineAll automount SAS MPIO VMFS ReFS Veeam Hyper-V

Continue through the project inventory