Have you heard of the 3-2-1 backup strategy? It’s like the ultimate safety net for your valuable data. You make three copies of your data, store them on at least two different storage media types, and keep one backup copy offsite. This strategy is very reliable and ensures that even if your hardware fails or some unexpected disaster strikes, you'll still have a backup of your data.
In this blog, we’ll provide an overview of how to implement this strategy using the SnapCenter Plug-in for VMware vSphere (SCV) and BlueXP backup and recovery for Virtual Machines.
Let’s first consider the storage that your on-premises data resides on. With the 3-2-1 backup strategy your data will be protected on both a primary and secondary ONTAP cluster. NetApp provides a lot of options so let’s take a quick look.
NetApp ONTAP Data Storage
ONTAP is NetApp’s industry leading storage solution that offers unified storage whether you access over SAN or NAS protocols. The 3-2-1 backup strategy ensures your on-premises data is protected on more than one media type and NetApp offers platforms ranging from high-speed flash to lower-cost media.
For more information on all of NetApp’s hardware platform’s check out NetApp Data Storage.
Overview and Architecture
Those familiar with the SnapCenter Plugin for VMware vSphere know the value of having a tightly integrated data protection offering that allows easy management of backup and restores for virtual machines. As part of the solution, SnapMirror provides a fast and reliable method to create a second immutable backup copy of virtual machine data on a secondary ONTAP storage cluster. With this architecture in place, virtual machine restore operations can easily be initiated from either the primary or secondary backup locations.
SnapCenter Plug-in for VMware vSphere documentation
Part of the NetApp BlueXP backup and recovery suite is a feature that integrates with the SnapCenter Plugin for VMware vSphere (on-premises) to extend a copy of the data to object storage in the cloud. This establishes a third copy of the data offsite sourced from the primary or secondary storage backups. BlueXP backup and recovery makes it easy to set up storage policies that transfer copies of your data from either of these two on-prem locations.
BlueXP backup and recovery documentation
For the backups transferred to object storage three major cloud providers are supported today.
In addition, you can use NetApp StorageGRID as the object storage backup target. For more on StorageGRID refer to the StorageGRID landing page.
With the most recent release of BlueXP backup and recovery, VMFS datastores are now supported in addition to NFS. The process for restores is a little different in that the VMFS datastore is mounted separately for the restore. However, all of this is automated and takes place seamlessly behind the scenes. Expanding our feature set is a primary focus at NetApp, so stay tuned for new features on an ongoing basis.
Choosing between the primary and secondary backups as the source in BlueXP backup and recovery will result in one of two topologies being implemented:
Fan-out Topology – When a backup is initiated by the SnapCenter Plug-in for VMware vSphere, a local snapshot is immediately taken. SCV then initiates a SnapMirror operation that replicates the most recent snapshot to the Secondary ONTAP cluster. In BlueXP backup and recovery,, a policy specifies the primary ONTAP cluster as the source for a snapshot copy of the data to be transferred to object storage in your cloud provider of choice.
Cascading Topology – Creating the primary and secondary data copies using SCV is identical to the fan-out topology mentioned above. However, this time a policy is created in BlueXP backup and recovery specifying that the backup to object storage will originate from the secondary ONTAP cluster.
NOTE: The choice of topology is available for each datastore that is chosen for backup.
The following architecture diagram provides a little deeper view of what’s going on under the hood:
Now let’s look at how to configure this solution so we can take advantage of this powerful strategy.
Establish a SnapMirror relationship between ONTAP clusters
For SnapCenter Plug-in for VMware vSphere to manage the transport of secondary SnapMirror and/or SnapVault copies to a second ONTAP Cluster, there are a few pre-requisites that need to be in place. While we won’t cover all the details, here are the steps for setting up SnapMirror:
- A cluster peer relationship between the ONTAP clusters must be established. This can be performed with System Manager or with the CLI. Refer to the ONTAP documentation for details.
- Next, peering is established between Storage Virtual Machines (SVMs) on each ONTAP cluster.
- Finally, a SnapMirror Data Relationship is established between volumes to be protected.
When creating a policy in the SnapCenter Plug-in you will see options for SnapMirror and SnapVault. If you choose SnapMirror, the retention schedule specified in the policy will be the same for both the primary and secondary snapshots. If you choose SnapVault, the retention schedule for the secondary snapshot will be based on a separate schedule implemented with the SnapMirror relationship. This is useful when you wish longer retention periods for secondary backups.
Configure the SnapCenter Plug-in for VMware vSphere
Once installed, the SnapCenter Plug-in for VMware vSphere can be accessed from the vCenter Server Appliance Management interface.
To configure backups of your virtual machines and datastores the following steps will need to be completed from the plug-in interface.
1.Discover the ONTAP storage clusters to be used for both primary and secondary backups.
2. Create backup policies specifying the retention period, frequency, and replication options.
NOTE: Snapshot labels are useful in that they can be used to enact policies with a specific retention period for the SnapVault copies replicated to the secondary ONTAP cluster. When SCV is used with BlueXP Backup and Restore, the Snapshot label field must either be blank or match the label specified in the BlueXP backup policy.
3. Create Resource Groups that contain the resources to be protected, the policy to be used, and the schedule.
Choose the datastores and/or virtual machines to be included in the backup:
NOTE: You have the option to select specific VMs or entire datastores. Regardless of which you choose, the entire volume (and datastore) is backed up since the backup is the result of taking a snapshot of the underlying volume. In most cases, it is easiest to choose the entire datastore. However, if you wish to limit the list of available VMs when restoring, you can choose only a subset of VMs for backup.
Choose the policy to be implemented for the backup:
NOTE: When using SCV in conjunction with BlueXP backup and recovery, it is only supported to use a single policy with each Resource Group.
Configure a schedule for the days and times that the backup will occur:
4. Go ahead and run your first backup. Once this is complete you are ready to log into BlueXP and set up the backups to take place from your on-premises ONTAP clusters to object storage on your cloud provider of choice.
NOTE: You must successfully run at least one backup before you will be able to discover your resources from BlueXP Backup and Recovery for Virtual Machines.
Configure Backups to Object Storage in BlueXP backup and recovery
For this section, it is a pre-requisite to have a Connector installed in BlueXP for the cloud provider that contains the object store you are backing up to.
Configuring backups of your datastores and virtual machines in BlueXP consists of the following steps:
1. Add your Working Environments, which in our case, will be our primary and secondary ONTAP storage clusters. Start by navigating to Storage Canvas and click on Add Working Environment.
From here click on Discover next to On-Premises ONTAP and fill out the credentials to add your system.
Repeat this process for each ONTAP cluster.
2. Navigate to Protection > Backup and recovery > Virtual Machines and click on Settings > SnapCenter Plug-in for VMware vSphere. Click on the Register button. This is where you will enter the SnapCenter Plug-in for VMware vSphere IP address and port number and your vCenter credentials.
Once registered a discovery process kicks off and you will see the datastore and virtual machine resources listed.
NOTE: Resources will only be discovered once a successful backup job has been run with the SnapCenter Plug-in. BlueXP only supports backing up resources scheduled with daily, weekly or monthly jobs. Resources associated with hourly backups will not be discovered.
3. Create policies for data retention. Navigate to Protection > Backup and recovery> Virtual Machines and click on Settings > Policies. Click on the Create policy and choose the retention schedule, backup source (primary or secondary on-premises backups), and archival policy. BlueXP Backup and Recovery supports tiering of backups to archival storage after a specified number of days.
4. We are now ready to start protecting our datastores and virtual machines. Back at the Protection > Backup and recovery main screen select the settings for a single datastore and select Activate Backup.
Follow the prompts to select a policy and next confirm that your Working Environment is available and ready. You have the option of adding an on-premises ONTAP cluster here if the Working Environment had not been previously added.
On the Select a Provider screen choose the cloud provider hosting the object storage service that you will be backing up to. In this example we will be showing you the configuration screen for Amazon Web Services.
On the Configure Provider page fill out the AWS key and secret and the region. This page also allows you to choose the archival tier for data that ages out based on the policy configuration.
Finally review the configuration and select Activate Backup to start protecting the datastores.
At this point, no data is immediately transferred. BlueXP backup and recovery checks with SCV every hour to see if any outstanding backups have been taken based on the schedule in SCV. Any outstanding snapshots are then copied to object storage and will be managed based on the retention schedule specified in the BlueXP backup and recovery policy.
Now we have a full 3-2-1 backup strategy implemented and can rest easy that our data has the most comprehensive protection available!
Restoring Virtual Machines in the case of data loss
Protecting your data is half the picture but being able to quickly restore data from any location when data loss or a ransomware attack occurs is critical to keeping your business running smoothly and meeting recovery point objectives. NetApp’s flexible 3-2-1 strategy allows for customized control of retention schedules at the primary, secondary and object storage locations. In this final section, we will briefly overview the process for restoring data from the SnapCenter Plug-in for VMware vSphere and from BlueXP backup and recovery.
The vCenter Plug-in for VMware vSphere includes the ability to do virtual machine and VMDK restores, as well as , Guest File Restores. There are a couple ways to access these features and perform a restore.
To restore a virtual machine, right-click on the VM in vCenter and go to SnapCenter Plug-in for VMware vSphere > Restore.
Alternatively you can navigate to a Datastore, click the Storage menu option, and then click Backups in the SnapCenter Plug-in for VMware vSphere section. Here you select an available backup and then items to be restored.
In the Restore wizard, choose the VM or specific VMDK to restore and whether to restore to the original or alternate location. On the Select location screen, you have the option to restore from the Primary or Secondary backup locations.
Review the Summary page and click on Finish to start the restore process.
BlueXP Backup and Recovery provides the capability of restoring virtual machines from object storage. Currently restoring to the original location is supported.
In BlueXP navigate to Protection > Backup and recovery and from the main page click on Virtual Machines to view the list of protected virtual machines. From the settings of the VM you wish to restore, click on Restore.
Follow the prompts to select the Backup….
Review the Summary and click on Restore.
Review the restore process from the Job Monitoring menu page.
Conclusion
We’ve taken you through the primary steps involved in implementing a comprehensive strategy for backing up virtual machines with the SnapCenter Plug-in for VMware vSphere and BlueXP backup and recovery for Virtual Machines. It’s a pretty easy ride once you get the hang of it. Control of retention schedules at each location allows for a lot of flexibility in designing solutions for disaster recovery and archiving. We’ll leave it at that for now, but please refer to the following resources for more information on this solution and its implementation.
NetApp BlueXP – NetApp-native backup and recovery for ONTAP workloads
BlueXP backup and recovery documentation
SnapCenter Plug-in for VMware vSphere documentation