Understanding vGW Series Fault Tolerance Support

This topic contains the following sections:

About vGW Series Fault-Tolerance

In the virtualized environment, fault-tolerance (FT) ensures continuous support of a virtual machine (VM) in the event of failure of the host on which it resides.

When you enable FT on a VM within VMware vCenter, a copy of the VM, called the secondary VM, is created automatically on another host. The original VM, referred to as the primary VM, and its copy, referred to as the secondary VM (VBM), run in lockstep. If the primary VM’s host fails, the secondary VM immediately assumes execution, without loss of connectivity, transactions, or data. For this to occur, the primary VM must be on a host that is part of a cluster of the same kind of hosts with the same configuration. Also, high availability must be enabled on the hosts comprising the cluster.

When you enable the FT feature, the secondary VM is created on a host that is either selected by DRS, if DRS is enabled, or is chosen from any available host in the cluster. The primary VM and the secondary VM have the same name and the same BIOS uuid, but each one has its own vc_uuid and vi_id.

The secondary VM has its own .vmx file. Both the primary VM’s .vmx file and the secondary VM’s .vmx file reside in the same data store directory.

When the primary VM’s host fails and the secondary VM takes control, from an external viewpoint it appears as if vMotion had moved the primary VM to the host of the secondary VM and the reverse, that is, as if the secondary VM was moved to the host where the primary VM resided.

vGW Series Fault Tolerance in the vGW Series

This section explains how the vGW Series handles VMs for which FT is enabled in the vCenter, and how it supports FT overall.

The vGW Series handles exposure of FT-enabled VMs to the user in the following ways:

vGW is prohibited from reconnecting vNICs and automatically suspending or resuming the VM. To do so would produce undesirable effects. For this reason:

For a VM that has been VMsafe secured for which FT has been enabled, the secondary VM will be created and its VMsafe param0 will be incorrect since it reflects the VC_uuid of the primary VM rather than its own. However, the vCenter will not try to reconfigure it, since the .vmx of the secondary is read-only and any reconfiguration operation will fail.

Enabling Fault Tolerance for a Virtual Machine

Before you enable FT for a VM, ensure that High Availability is enabled for the cluster.

To enabled FT for a VM in the vCenter:

  1. Use the vSphere client to access the vCenter, and locate the host where the VM resides.
  2. Right-click the name of the VM.
  3. From the displayed menu, select Fault Tolerance.
  4. Select Turn On Fault Tolerance.
  5. After reviewing the message noting that DRS automation will be disabled and that the memory reservation of the VM will be changed to the memory size of the VM, accept the changes and click Yes.

    You c

Verify in the Recent Tasks window at the bottom of the screen that fault tolerance was turned on for the VM.

Related Documentation