Error: Out of resources during Zerto VPG creation on 3-node vSAN cluster
search cancel

Error: Out of resources during Zerto VPG creation on 3-node vSAN cluster

book

Article ID: 449761

calendar_today

Updated On:

Products

VMware vSAN

Issue/Introduction

Zerto Virtual Protection Group (VPG) creation or journal provisioning fails on vSAN datastores when the default Zerto storage policy exceeds the physical host capacity and available fault domains of the cluster. This typically occurs in 3-node clusters where the policy defaults to a configuration requiring four or more fault domains. Users may see the vSAN datastore report "Out of resources" despite having sufficient raw storage space.

Symptoms

  • VPG creation fails during the "provisioning replication journal" phase.
  • vCenter tasks appear stalled or time out at 0% during disk creation.
  • vCenter tasks show the error: A general system error occurred: Error creating disk Out of resources.
  • Zerto logs or vCenter events report: There are currently 3 usable fault domains. The operation requires 1 more usable fault domains.
  • Log snippets include: Failed to execute Single CreateDisk on VM vm-#### with retries. Exception: VMBasicsException: Handle vmware fault Message: A general system error occurred: Error creating disk Out of resources.

Environment

  • VMware vSAN 
  • Zerto Virtual Replication
  • vSAN Cluster with 3 ESXi hosts

Cause

The "Zerto Default vSAN Policy" is configured to use RAID-5 (Erasure Coding) by default. In a vSAN environment, RAID-5 requires a minimum of four fault domains (hosts) to distribute data and parity components. In a 3-host cluster, vCenter cannot satisfy this placement requirement, leading to a resource error regardless of physical disk space.

Resolution

Modify the Zerto storage policy to align with the 3-node cluster limits:

  1. Log in to the vSphere Client.
  2. Navigate to Policies and Profiles > VM Storage Policies.
  3. Locate and edit the Zerto Default vSAN Policy.
  4. Under the vSAN tab, select Edit Storage Policy.
  5. Change the Failure tolerance method from RAID-5 (Erasure Coding) to RAID-1 (Mirroring) - restricted.
  6. Ensure Failures to Tolerate (FTT) is set to 1 Failure - RAID-1 (Mirroring).
  7. Save the changes and click OK.
  8. Wait for the policy status to show as Compliant for all associated objects.
  9. Retry the VPG creation or journal provisioning in the Zerto Analytics/Console.

 

Additional Information

  • For clusters with 4 or more hosts, RAID-5 is supported and remains the recommended policy for space efficiency.
    • Best practice is to have 5 or more hosts when using a RAID-5 policy in a production setting in order to account for routine maintenance and patching
  • This behavior is by design for RAID-5 on 3-node clusters.
  • See Download Broadcom Products and Software to download the latest Zerto or vSAN patches if automation updates are released.
  • Subscribe to this article to be updated on fix status or policy automation updates: How to subscribe to Knowledge Base articles.
  • If the issue persists after policy modification, verify host health using esxcli vsan cluster get to ensure no hosts are in a decommissioning or partitioned state.
Powered by Wolken Software