Migrating VMs from VMware to OpenStack with Pure Storage (Part 2)

This is Part 2 of our VMware to OpenStack migration series. If you haven’t read Part 1 yet, check out VMware to OpenStack: The Simple Way for the foundational concepts and general migration approach.

When you have Pure Storage FlashArray as your storage foundation, migrating VMs from VMware vSphere to OpenStack becomes significantly more streamlined. Building on the concepts from Part 1, this guide focuses specifically on leveraging Pure Storage’s advanced capabilities to make your migration even smoother.

Good news for OpenStack 2025.1 (Epoxy) users: If you’re running the latest OpenStack release, this is essentially a single-step process! The Pure Storage Cinder driver now fully supports volume groups, so you can skip most of the complexity described below.

For older OpenStack versions: You’ll need to follow the three-step approach outlined in this guide to work around limitations in earlier Cinder drivers.

Why This Migration Matters

Many organizations are looking to modernize their infrastructure by moving from traditional virtualization to more flexible cloud platforms like OpenStack. When Pure Storage serves both environments, you can take advantage of advanced storage features to make this transition smooth and efficient.

See the Migration in Action

To help you follow the process more easily, I’ve included a detailed video walkthrough of the migration from VMware (starting with a VMFS-based VM) to OpenStack using Pure Storage. This video shows the actual steps performed through the VMware vSphere and OpenStack GUIs, providing a clear visual guide alongside this blog’s instructions.

Watch the full migration video here:
VMware to OpenStack Migration Video (GUI)

The Migration Process: It Depends on Your Setup

The complexity of your migration depends on two factors: your current storage setup and your OpenStack version.

Already Using vVols + OpenStack 2025.1 (Epoxy) or Later: Single Step!

If your VMs are already running on Pure Storage vVols and you’re using OpenStack 2025.1 or newer – you’re in luck! Skip directly to Step 3 and import your volumes into OpenStack. That’s it!

Using VMFS Storage: Two or Three Steps

If your VMs are currently on traditional VMFS datastores, you’ll need to start with Step 1 to move them to vVols first, then:

  • OpenStack 2025.1+: Skip to Step 3 (two steps total)
  • Older OpenStack: Follow all three steps

Already Using vVols + Older OpenStack: Two Steps

If you’re already on vVols but running an older OpenStack version, skip Step 1 and proceed with Steps 2 and 3.

Step 1: Move Storage to vVols in vSphere (VMFS Users Only)

Skip this step if your VMs are already using Pure Storage vVols!

This first step is only needed if your VMs are currently running on traditional VMFS datastores. If you’re already using vVols, jump ahead to the appropriate step based on your OpenStack version.

What You Need:

  • Pure Storage FlashArray properly connected to vSphere
  • Enough space on your vVols datastore
  • VMs without active snapshots

The Process:

  1. Open vSphere Client and find your VM
  2. Right-click the VM and select “Migrate”
  3. Choose “Change storage only”
  4. Select your Pure Storage vVols datastore as the destination
  5. Pick the right storage policy for your needs
  6. Review your settings and start the migration
  7. Monitor the progress until complete

This step transforms how your VM’s storage is managed, setting it up for the next phase while keeping everything running smoothly.

Step 2: Clone Your Storage Volumes (Older OpenStack Versions Only)

Skip this step if you’re using OpenStack 2025.1 (Epoxy) or later!

For older versions of OpenStack, you need to create copies of your vVols volumes that OpenStack can understand. This happens on the Pure Storage array itself.

Best Practice Tip: Following vVols best practices, your volumes should be in non-replicated pods on the FlashArray. This setup allows you to keep the original VM vVols intact on the array as a recovery option – you’re essentially creating a migration copy while preserving your safety net.

The Process:

  1. Log into your Pure Storage FlashArray interface
  2. Navigate to Storage > Volumes
  3. Find your vVols volumes using your VM name
  4. Click on each volume and select “Copy”
  5. Give your new volume a clear name
  6. Complete the cloning process

The great news? Pure Storage’s cloning is instant, so this step completes immediately without moving any actual data.

Step 3: Import Volumes into OpenStack

The final step brings your storage volumes under OpenStack’s control through the Horizon dashboard.

The Process:

  1. Log into OpenStack Horizon with admin privileges
  2. Go to the Admin section and find Volume management
  3. Click “Manage Volume” to import existing storage
  4. Fill in the details:
    • Give it a name in OpenStack
    • Point to your Pure Storage volume
    • Set the correct volume type
    • Mark boot disks as bootable
  5. Submit the form and verify the volume appears

Key Benefits of This Approach

Minimal Downtime: Your VMs keep running throughout most of the process, reducing business disruption.

No Data Movement: Pure Storage’s advanced features mean you’re not actually copying large amounts of data across the network.

Preserved Performance: Your applications maintain the same high performance they had on Pure Storage in vSphere.

Safety First: Each step includes verification points to ensure everything is working correctly before proceeding. Plus, when following vVols best practices with non-replicated pods, your original VM vVols remain on the FlashArray as a backup recovery option.

Things to Keep in Mind

  • Plan your migration during maintenance windows when possible
  • Test the process with non-critical VMs first
  • Make sure you have good backups before starting
  • Document your VM configurations before migration
  • Monitor performance after the migration to ensure everything is working well

Common Gotchas

  • Remove VM snapshots before starting the migration
  • Double-check that your Pure Storage is properly configured for both environments
  • Verify network connectivity between all systems
  • Make sure you have enough storage space for the migration

Automate Everything with Ansible

Did you know? All of these manual GUI steps can be automated using Ansible playbooks! If you’re managing multiple VMs or want to ensure consistent, repeatable migrations, automation is the way to go.

From Storage vMotion operations in vSphere to volume cloning on Pure Storage and OpenStack volume management – there are Ansible modules for each step of this process. This means you can script your entire migration workflow, handle error conditions gracefully, and run migrations at scale.

Look for another post in this series that will dive deep into the Ansible automation approach for these migrations. We’ll show you how to turn this manual process into a fully automated workflow that you can run with a single command.

Wrapping Up

Migrating from vSphere to OpenStack doesn’t have to be a headache when you have Pure Storage as your foundation. This three-step process leverages Pure Storage’s capabilities to make the transition as smooth as possible while maintaining the performance and reliability your applications need.

The key is taking it one step at a time, verifying each stage before moving forward, and leveraging Pure Storage’s advanced features to do the heavy lifting for you. With proper planning and execution, you can modernize your infrastructure while keeping your applications running smoothly.


What’s Next?

Ready to take your VMware to OpenStack migration to the next level? In Part 3 of this series, we dive deep into automating the entire migration process with Ansible. Learn how to transform the manual steps outlined in this guide into a fully automated, repeatable workflow that can handle enterprise-scale migrations with confidence.

👉 Continue to Part 3: Automating VMware to OpenStack Migration with Ansible

In Part 3, you’ll discover:

  • Complete Ansible playbooks for each migration step
  • Error handling and rollback strategies
  • Scaling migrations across multiple VMs
  • Best practices for production deployments
  • Real-world examples and troubleshooting tips

Don’t miss out on turning this manual process into a one-command migration solution!

Leave a Reply

Your email address will not be published. Required fields are marked *