NVMe vVol Datastore Mounting Fails with InaccessibleDatastore Error
search cancel

NVMe vVol Datastore Mounting Fails with InaccessibleDatastore Error

book

Article ID: 439460

calendar_today

Updated On:

Products

VMware vSphere ESXi VMware vCenter Server

Issue/Introduction

  • When attempting to mount a new NVMe vVol datastore on an ESXi host via Fibre Channel, the operation fails with the following error in the vSphere Client:


vSphere HA failed to create a configuration vVol for this datastore and so will not be able to protect virtual machines on the datastore until the problem is resolved. Error: (vim.fault.InaccessibleDatastore)

  • Mounting the same datastore using SCSI vVol may work correctly on the same host and array.

  • In the /var/log/vvold.log on the ESXi host you may see errors indicating the missing PE or missing NQN for the array:
    • getContainerAccessibility, No NVMe subsystemNQN found for array
    • VvolFileSystemImpl: No PEs for array
  • Additionally, running localcli storage vvol storagecontainer list may show the container as Accessible: false. 

Environment

  • VMware ESXi 8.x
  • VMware vCenter 8.x

Cause

  • The host is unable to establish a connection because it is receiving no Protocol Endpoint (PE) information from the storage array for the NVMe subsystem.
  • This is due to a configuration issue on the storage array side.

 

Resolution

  • This issue must be resolved through storage array configuration. Coordinate with your storage vendor to verify the following:
    1. NVMe Namespaces: Ensure that NVMe namespaces are correctly configured and available.
    2. Protocol Endpoint (PE): Verify that the PE is available and correctly presented to the ESXi host.
    3. Container Configuration: Confirm the Storage Container is explicitly enabled for NVMe.
    4. Port Configuration: Ensure storage ports are configured to allow NVMeoFC.
    5. Persona and Mapping: Verify the host persona, NQN mapping, and zoning are correct for NVMe over Fibre Channel.

Additional Information