You can use the vCloud Availability solution to support a replication from a vCloud Director site to a second vCenter Server.
You can make use of this capability when your original vCenter Server is offline and you must migrate the VM from the vCloud Director site to another vCenter Server. This use case is often called Unplanned Failover. For more information about the replication topology, see vCloud Availability for vCloud Director Portal Overview in the vCloud Availability for vCloud Director Installation and Configuration Guide.
Verify that the virtual machine is successfully recovered in the vCloud Director site through a Failover task. You can perform a Failover task using the vCloud Availability Portal. For more information, see Perform a Failover Task Using the vCloud Availability Portal .
Verify that the virtual machine is detached and does not appear in the Workloads tab of the vCloud Availability Portal. You can detach a workload by running a Detach task in the vCloud Availability Portal. For more information about detaching workloads through the vCloud Availability Portal, see Detach a Virtual Machine Using the vCloud Availability Portal. .
If you are recreating the second vCenter Server on top of the original on-premise environment, for example, reconnecting the same ESXi hosts, verify that you do not need the original on-premise VM and delete it from the inventory to avoid a UUID conflict.
- Use the vSphere Web Client to connect to your second vCenter Server.
- Navigate to vSphere Replication tab under Monitor, and click Incoming Replications.
- Above the list of incoming replications, click the Configure replication from cloud provider icon ().
The Configure Replication From Cloud Provider wizard opens.
- On the Source site page, select the cloud provider site where the virtual machine is located.
If you have created a connection to the cloud provider, select the source virtual data center from the list and click Next.
If the status of the connection is Not authenticated, you must provide credentials to authenticate with the cloud organization.
If you have not created a connection to the cloud provider, click New Provider VDC, click Next, and follow the on-screen prompts to connect to the target cloud organization.
- On the Available virtual machines page, select the virtual machine that you want to replicate.
You can select only one virtual machine from a vApp.
- Accept the automatic assignment of a vSphere Replication server or select a particular server on the local site and click Next.
- On the Target location page, click Edit to select the datastore where replication data is saved.
If you want to use existing disks as seeds for the replication, browse the datastore to locate the folder where the seed disks are located.
- (Optional) To configure a replication of individual disks, click the name of the source virtual machine.
The list of disks on the source virtual machine expands.
For each disk, you can select the virtual format, storage policy, and a datastore where it is replicated. If the source virtual machine contains more than one disk, you can disable the replication of a disk by clicking Disable in its Replication Enabled row.
- (Optional) On the Replication options page, select the quiescing method for the guest OS of the source virtual machine.
Quiescing options are available only for VMs that support quiescing. vSphere Replication does not support VSS quiescing on Virtual Volumes.
- (Optional) Select Network Compression.
Compressing the replication data that is transferred through the network saves network bandwidth. Compressing the replication data might also help reduce the amount of buffer memory used on the vSphere Replication server. However, compressing and decompressing data requires more CPU resources on both the source site and the server that manages the target datastore.
- (Optional) On the Failback recovery settings page, select the Configure recovery settings check box, and then select a virtual machine folder and host or resource pool.
- On the Recovery settings page, use the RPO slider or the time spinners to set the acceptable period for which data loss is acceptable in the case of a site failure.
The available RPO range is from 15 minutes to 24 hours.
- (Optional) To save multiple replication instances that can be converted to snapshots of the source virtual machine during recovery, select Enable in the Point in time instances pane, and adjust the number of instances to keep.
- On the Ready to complete page, review the replication settings, and click Finish.
A virtual machine configuration task appears in the Recent Tasks list at the bottom of the vSphere Web Client. A progress bar indicates that the source virtual machine is being configured for replication.
If the configuration operation completes successfully, the replication task that you created appears in the list of incoming replications on the vSphere Replication tab under Monitor in the vSphere Web Client.
The virtual machine appears in the Reversed tab of the vCloud Availability Portal.
The virtual machine is running in the vCloud Director (cloud) site and is protected on the second vCenter Server (on-premise) site.