How vSAN handles Policy Changes between (RAID0/ RAID1 Mirroring) and RAID 5/6 Storage Policies.

Products

• VMware vSAN

Issue/Introduction

When modifying storage policies between RAID 0/1 (mirroring) and RAID 5/6 (erasure coding), vSAN must re-layout the objects to comply with the new policy. This process generates transient capacity consumption. If too many objects are reconfigured simultaneously—particularly large ones—the vSAN datastore may reach full capacity, potentially causing cluster-wide impact.

Capacity Planning Example
Consider a virtual machine (VM) with 100 GB on-disk size:

RAID 1 (FTT=2): Requires 300% of the allocated space (300 GB).

RAID 6: Requires 150% of the allocated space (150 GB).
During the transition from an FTT=2 (RAID 1) policy to a RAID 6 policy, the system may require up to 450 GB of transient capacity to complete the re-layout of components. VMware by Broadcom recommends applying storage policy changes in small batches to avoid exhausting datastore capacity during this process.

Environment

VMware vSAN(All versions)

Resolution

When migrating between policy types, be aware of the following operational impacts:

Resynchronization Traffic: The system performs an object re-layout, which generates significant backend network traffic and internal I/O.
Performance Degradation: ESXi hosts may experience increased CPU utilization and vSAN network latency during the migration.
Backup Job Contention: Backup solutions (e.g., Cohesity, Veeam) may compete for resources. Pause all non-essential backup and maintenance jobs before initiating policy changes.

Resolution and Best Practices
To prevent capacity issues, verify available free space on the vSAN datastore before starting and apply policy changes in small batches.
If insufficient transient space is available to transition to a more efficient policy:

• Migrate the VM off the vSAN datastore, apply the new policy, and migrate it back.
• If the VM currently uses a RAID 1 policy, temporarily switch to FTT=0 (RAID 0) to remove mirror copies and free capacity.

Warning: This removes all data redundancy. Any hardware or host failure affecting the remaining active component will result in permanent data loss. Use this option only as a temporary measure.

Migration Procedure

1. Navigate to the vSphere Client.
2. Select the target RAID 5 or RAID 6 policy under VM Storage Policies.
3. Right-click the target VM, select VM Storage Policies, and choose Edit VM Storage Policies.
4. Apply the new policy to the desired VMDK files.
5. Monitor the progress in the vSAN Monitor dashboard under the Resyncing Objects tab.

Additional Information

Architecture:

• Design and Operational Guidance for vSAN Storage Clusters 
• What are vSAN Policies
• Storage space utilization calculation for VM's objects on vSAN Express Storage Architecture (ESA) datastore
• Using RAID 5 or RAID 6 Erasure Coding in vSAN Cluster
  
  

Monitor:

• Monitor vSAN Capacity
• vSAN Health Service - Capacity utilization - Storage space
• vSAN Health Service - Cluster health - vSAN optimal datastore default policy configuration
  
  

How To:

• Change the vSAN storage policy on VMs to save space