Introduction

Replication is the process of copy a VM from one host or cluster to another. Replication is the easiest way to move a VM from your on-prem environment to your virtual data center.

This guide is describing a one-way migration to your vDC, it does not cover a disaster recovery scenario with an emergency fail-over with network migration or fail-back to an on-prem environment.

Info

This guide is based on the premise that Veeam Backup & Replication is installed and configured with a VMware vCenter or vSphere ESXi host as source for the on-prem VMs. Please refer to Veeam for installation and configuration.

We also recommend using the latest version of Veeam.

Procedure

Prepare target virtual data center (vDC)

1. Create a Portal account.
   Veeam is using the resource of a VMware Cloud Director organization to store the replicated VM directly in the specified vDC (virtual data center). This means that an organization must be present when Veeam on the Service Providers side is configured.

2. Create a vDC.
   A vDC is the target for the replication, therefor at least one vDC must be created in the organization.

3. Configure network as needed (firewall, VPN tunnel, internal network etc) for the vDC and optionally create a Virtual Machine to test network connectivity.

4. Notify Cloudist Solutions support@cloudist.se that you want to replicate VMs and inform us which organization this concerns. Once we have prepared your organization, we will notify you, so that you may continue with the rest of the setup.

5. Create a user for Veeam. In order to authenticate Veeam against VMware Cloud Director, a user with Organization Administrator role in VMware Cloud Director must be used.

Prepare Veeam 

1. Start the Veeam Backup & Replication Console and navigate to BACKUP INFRASTRUCTURE and select the Service Provider option. Click "Add Provider" in the upper left corner.

   
   
   
2. Type in "cc01.fortlax.se" as DNS name or IP address and select Next.
   
   
   
3. Add credentials for the user that was prepared in Prepare vCD step 5.
   
   
   
4. Review the Replica Resources to make sure that the correct vDC is showing. If not, it might be that the desired vDC has been added after the tenant was created on the Service Provider side. If a vDC is missing, contact support@cloudistsolutions.com to resolve the issue.
   
   
   
5. After applying the "Replica Resources" Veeam updates its internal database with information about the resources available from the Service Provider.
   
   
   
6. Finish the guide to save the Service Provider. Veeam will initiate a rescan to gather all necessary information from the Service Provider.
   
   
   

Considerations

Steps

1. Create new replication job of virtual machine
   
   
   

2. Give the replication job a descriptive name. There is no need to select any of the options as this is a one-way replication and not a disaster recovery setup.
   
   
   
3. Select the Virtual Machines that will be replicated in this job. For this type of migration, source should be configured as "From production storage (actual VM state)" otherwise synchronization before fail over will not take into account changes made to the source after last backup.
   Multiple jobs will make partial fail over easier as it will give you greater control over what Virtual Machines will be failed over. There respective jobs can be disabled in order to protect the destination Virtual Machine from accidental update from the source environment. It will also give fewer warnings, which will help with monitoring.
   
   
   
4. Select "Cloud host..." as destination.
   
   
   
5. Select the target vDC for the job.
   
   
   
6. Select the desired vApp as target, if there are no vApps in the vDC, Veeam will create a default vApp named "Cloud Connect 1". There is also a setting for Storage policy that will be used for the Virtual Machines in this job.
   
   
   
7. Clear the "Replica name suffix" field in order to keep the Virtual Machine name intact after replication. As this is no regular backup or disaster recovery scenario, there is no need to keep more than one restore point.
   
   
   
8. Leave the Data Transfer settings to default.
   
   
   
9. Leave the Guest Processing settings to default. Final synchronization will be performed after a graceful shutdown of the source virtual machine has been made.
   
   
   

10. It is recommended to run the replication on a schedule. This will make the subsequent synchronizations smaller.
    
    

    As initial replication can take some time, plan ahead and make sure that the initial replication will finish well before planned fail over.

    
    
    
    

11. Review the Summary and Finish the setup. There is an option to run the job for the first time directly after finishing the setup guide.
    
    
    

12. The job will show up under HOME/Jobs/Replication in the Veeam Backup & Replication Console.

    The duration of this operation depends on the total hard disk size of the virtual machine, network connection bandwidth and network latency to target environment.

    

13. After the job has been run, review the result to verify that it has been successful, both in Veeam Backup & Replication Console and in the target vDC.
    
    
    
    
    

14. Once a replication job has finished successfully, it is possible to power on the target VM.

    Powering on the target VM while the source VM is still powered on, might result in unexpected or unwanted behaviour depending on your networking setup. 

Failing over a VM

Failing over is the process of moving the workload from the on-premise environment to the target virtual data center. This will require a maintenance window for the VM.

1. In order to do a final synchronization, the source VM must be powered off. Otherwise application consistency can not be guaranteed and changes made to the source after last synchronization will be lost.

2. Run previously created replication job in Veeam Backup & Replication Console.

3. After the replication job is complete, review the settings of the target VM, especially network settings. It is recommended to configure "Hot Add" for vCPU and vMem. These settings can only be changed when the VM is powered off.

4. (Optional) Disable the scheduled replication job in Veeam. If the target VM is powered off after failing over the workload, replication will overwrite the target. If the target is powered on, replication will fail with errors.

5. Power on the target VM and make sure network connectivity is working. Make sure any on-prem dependencies are updated (DNS records, IP addresses, firewall rules etc).

6. After verifying function of the target VM, the replication job can be disabled, provided it does not handle VMs that are yet not failed over.

7. (Optional) Set desired backup retention for the VM in the portal.