jules - Fotolia

Try an Azure Site Recovery setup for DR needs

Administrators who need some assurance that mission-critical workloads will stay up when trouble strikes can try out Microsoft's cloud-based disaster recovery service.

In today's IT world, you can have workloads on premises and in the cloud. One common denominator for each location is a need to plan for disaster recovery. Azure Site Recovery is one option for administrators who need a way to cover every scenario.

Azure Site Recovery is a service used to protect physical and virtual Windows or Linux workloads outside of your primary data center and its traditional on-premises backup system. During the Azure Site Recovery setup process, you can choose either Azure or another data center for the replication target. In the event of a disaster, such as a power outage or hardware failure, your apps can continue to operate in the Azure cloud to minimize downtime. Azure Site Recovery also supports cloud failover of both VMware and Hyper-V virtual infrastructures.

One of the real advantages of this Azure service for a Windows shop is integration. All the functionality is built right into the admin portal and requires little effort to configure beyond the agent installation, which can be done automatically. Offerings from other vendors, such as Zerto and Veeam work the same way but require additional configuration using a management suite based outside the Azure portal.

Azure Site Recovery pricing

One of the big issues for any platform is cost. Each protected instance costs $25 per month with additional fees for the Azure Site Recovery license, storage in Azure, storage transactions and outbound data transfer. Organizations interested in testing the service can use it for free for the first 31 days.

As with most systems, there are caveats, including how replication and recovery are tied to specific Azure regions depending on the location of the cluster. There is a list of supported configurations on Microsoft's documentation.

Azure includes the option to failover to an on-premises location, which reduces the cost to $16 per instance. However, this option requires meeting bandwidth requirements that are not a factor in an Azure-to-Azure failover scenario.

Azure Site Recovery uses vaults to store workload dependencies

Most disaster recovery (DR) environments utilize the concept of crash consistent applications, meaning the application fails over as a whole with all its dependencies. In Azure, you store the VM backups, their respective recovery points and the backup policies in a vault.

These vaults should contain all the servers that make up the services required for a successful failover. (You should test before an emergency occurs to make sure it functions expected.) It is possible to fail over individual VMs within a replication group if needed; this used to be an all-or-nothing scenario until recently.

How to create a Recovery Services vault

For this Azure Site Recovery setup tutorial, we'll cover how to configure VMs for site-to-site replication between regions via the portal.azure.com link.

As with most Azure tools, the Disaster Recovery menu is on the left-hand side with the other Azure services. Under this menu is the Recovery Service vault option. Create one by filling in the fields as shown in Figure 1.

Recovery Services vault
Figure 1. Create the Recovery Services vault by filling in the project details.

When you have entered all your specifications, click Create to build the vault. The next step is to choose the purpose for the vault. The choice is either for backup or DR.

Next, add the VMs. To start, from the vault choices select Site Recovery.

From the on-premises option, click Replicate Application to open a wizard to add VMs. Next, click Review + Start replication to start the creation and replication process, which can take several minutes. For ease of access and experimentation purposes, I suggest pinning it to your dashboard. Opening the vault provides a health overview of the site and clicking on each item shows details about the replication status as shown in Figure 2.

Azure Site Recovery replication
Figure 2. The Azure Site Recovery setup process replicates the protected instances to the defined target, either an on-premises location or in Azure.

This completes the creation of a group with two protected VMs. Every VM added to that resource group automatically becomes a protected member of the vault. By default, the DR failover is set to a maximum duration of 24 hours. After the initial configuration, you can adjust the failover duration and snapshot frequency from the Site Recovery Policy - Retention policies page.

The last step is to create a recovery plan. From the vault, select Create recovery plan and then + Recovery plan and give it a name as shown in Figure 3 with our example called MyApplicationRecoveryPlan. You choose the source from either the on-premises location or Azure, and the Azure target.

Azure Site Recovery plans
Figure 3. The menu displays all your configured recovery plans in Azure Site Recovery. From here, you can execute a test failover to verify your settings.

When complete, opening the plan to verify it works properly by clicking Test for a nondisruptive assessment that checks the replication in an isolated environment. This process can detect any problems related to services and connectivity the application needs to function in a failover setting.

This tutorial covers some of the basic functionality of Azure Site Recovery. For more granular control, there are many more options available to provide advanced functionality.

Dig Deeper on IT operations and infrastructure management

Cloud Computing
Enterprise Desktop
Virtual Desktop