Microsoft recently made available a preview service, Hyper-V Recovery Manager (HVRM). This is a hybrid service that leverages the high ground of the Azure public cloud in a smart way. HRVM manages the replication of your primary data center to your disaster recovery (DR) site. HVRM is a service for Windows Azure customers that appears in the Windows Azure Customer Portal after you sign up for the service. From the portal, you can perform emergency, on-demand, and test failovers of your infrastructure to the DR site. You can achieve excellent DR readiness protecting virtual machine (VM) workloads at a fraction of the cost compared to traditional approaches.While making clear that this is a preview service and all features are subject to change, the HVRM service itself is a first, and powerful, example of a future vision becoming reality for Microsoft. That is, the use of the Azure public cloud for off-premise automation controlling on-premise private cloud fabric. Microsoft is working towards a future where the entire network smarts for an organization can live in Windows Azure and/or Microsoft partner clouds. In this future, on-premise and private cloud-based compute, storage, and networking resources are remotely provisioned as fabric commodities using a cloud-based management engine.
Readers familiar with the Windows InTune service for cloud-based PC management might see this as the beginning of an "InTune for Datacenters". Services are going to migrate to where they can be most economically delivered from. The one-to-many scale reachable from Windows Azure global data centers gives Microsoft a delivery vehicle for this new paradigm of massively scalable network management from the cloud. When combined with the groundbreaking Hyper-V Replica feature in Windows Server 2012, organizations can have excellent DR readiness at a fraction of the cost compared to all previous data center DR architectures.
Hyper-V Recovery Manager (Preview)
The customer-premise technology that makes HVRM possible is Hyper-V Replica, a technology native to Windows Server 2012. So to leverage HVRM, you need to standardize on Hyper-V as your data center hypervisor. In addition, you need to be using System Center 2012 Service Pack 1 (SP1), in particular System Center 2012 Virtual Machine Manager SP1 (SCVMM) as your virtualization management software.Azure-based automation talking to on-premise SCVMM, which is in turn driving Hyper-V Replica, is how HRVM works. Figure A shows the arrangement of the basic components of the service; notice the portal hosted in Azure, which is the ‘always on' DR dashboard for protected resources. The solution uses familiar tools: Windows Server 2012, System Center 2012 SP1, and the Windows Azure customer portal.
Topology of Hyper-V Recovery Manager is simple and uses familiar tools.Figure B is a full-size screen shot of the Azure portal at the top left of Figure A. In Figure B, the Recovery Plan view is seen-observe the Failover and Test Failover buttons are easy to find and intuitive to use. These functions are found in the Windows Server 2012 Hyper-V management console (of each Hyper-V replica server) on a per-VM basis. HVRM exercises the failover and test failover functions automatically and for complete clouds. All VM actions in cloud failover and test tasks are orchestrated by HVRM in Windows Azure.
The Recovery Plan page in Windows Azure Recovery Services, Azure portal
Not only is HVRM a "game changer" in the discipline of Disaster Recover (DR) but in the smaller and mid-range spaces, SAN to SAN replication for the purpose of protecting virtual machines becomes an expensive and legacy technology. Why invest in a particular storage vendor's proprietary SAN to SAN hardware replication when Hyper-V replica can accomplish the same objective (a "hot DR site" with up-to-date, ready to start VM copies) at a fraction of the cost?
Complete steps to deploy HVRM preview
To try the HVRM feature in preview, you will need a Windows Azure account. The Hyper-V Recovery Manager limited preview program is available to a small group of customers using Windows Server 2012 and System Center 2012 SP1. For the Hyper-V Recovery Manager limited private preview, only customers located in the following countries will be considered: United States, Canada, United Kingdom, Germany, France, Belgium, Switzerland, Denmark, Netherlands, Finland, Australia, Japan, India, and New Zealand. If you have the Windows Server 2012 and System Center 2012 SP1 infrastructure to test HVRM, you can apply for the trial at this link.
If you are selected to participate in the trial, here are all the steps to follow to enable DR protection:
- A welcome email will be received.
- Configure a new storage vault in Azure using the provided link.
- A new Recovery Services service tab appears in the Azure portal after creating the vault.
- Clicking the Recovery Services icon for the first time will list the steps to get HRVM setup (See Figure C for the Quick Start steps):
- Upload an X.509 certificate, this can be a personal User certificate issued by private CA.
- Download the HRVM agent and run setup on the production data center instance of SCVMM.
- Import the X.509 private key PFX to the computer store of the SCVMM computer. Then select it at the certificate registration prompt.
- The HVRM vault you created in step 2 is found automatically in the cloud by searching on the thumbprint of the private certificate.
Recovery Services Quick Start steps to get Hyper-V Recovery Manager setup.
- Repeat HVRM provider setup (under step 4) on the DR site SCVMM instance.
- HVRM requires clouds to exist at the source and target sites: DR is managed at the Cloud level. If necessary, move protected VMs into a new cloud at the production site. If necessary create a cloud with matching features at the DR site. Clouds are be automatically detected by Azure Recovery Services.
- After creating a target cloud at the DR site, replication settings can be set in Azure.
- Once the cloud has protected status in Azure, the Enable Recovery setting becomes available in the SCVMM console. Enable recovery for all desired VMs in that cloud.
- If you have created Hyper-V Replicas already, you are warned that the new replication settings from Azure will replace the previous settings from Hyper-V.
- Existing Hyper-V replicas in the DR site are moved into the DR cloud by the Azure provider.
- Azure identifies the DR cloud as the failover cloud for the production cloud.
- Same-named virtual networks in both clouds are auto-mapped.
- A top-level dashboard in the Azure portal shows how many VMs are protected, and has links to download the SCVMM HVRM agent and upload X.509 certificate(s).
- After VMs are protected, you can create a Recovery Plan.
- A recovery plan has groups and steps that you can customize and re-order as desired.
- Recovery plan groups let you segregate some VMs in a cloud to failover before or after other groups. You can add scripts (PS1 files that reside in file shares on VMM servers) and manual actions, and order the scripts, actions, and groups.
- On-premise Hyper-V management console continues to accurately report on replica status. Pre-existing replicas are used by Azure HVRM and not re-created or re-replicated.
John Joyner, MCSE, CMSP, MVP Cloud and Datacenter Management, is senior architect at ClearPointe, a cloud provider of systems management services. He is co-author of the "System Center Operations Manager: Unleashed" book series from Sams Publishing, and is developing cloud-based management solutions based on the Microsoft System Center 2012 suite. John is a retired U.S. Navy Lt. Commander 'Surface Warfare Officer', with the subspeciality 'Computer Scientist, Proven'. His tours of duty included Chief of Network Operations for NATO's southern region and network administrator aboard the aircraft carrier USS CARL VINSON (CVN-70).