StarWind is a hyperconverged (HCI) vendor with focus on Enterprise ROBO, SMB & Edge

Deep dive into VMware vCenter Site Recovery Manager 8.1. Part 1 – Setting up the solution

  • March 8, 2019
  • 11 min read
Cloud and Virtualization Architect. Kevin focuses on VMware technologies and has vast expertise in cloud solutions, virtualization, storage, networking, and IT infrastructure administration.
Cloud and Virtualization Architect. Kevin focuses on VMware technologies and has vast expertise in cloud solutions, virtualization, storage, networking, and IT infrastructure administration.


I think that any business needs a disaster recovery site. That’s why I discuss today VMware vCenter Site Recover Manager (SRM) 8.1 and how to set up that solution.


vCenter Site Recovery Manager in a nutshell

VMware vCenter Site Recovery Manager has been introduced back in 2011. It is a VMware vCenter plug-in for disaster recovery site configuration and management. It also allows migrating to that site at the moment of need or during planned migrations. In other words, SRM ensures shortest services downtime if something goes wrong at your main site. Please, don’t count on that thing that much since there will be insignificant time losses anyway.

This solution utilizes VMware vSphere Replication mechanism to automate VM migration to a disaster recovery site.

SRM implements 3 approaches towards VM replication:

  • Datastore Groups – VM datastore group array-based replication.
  • Individual VMs – replicating VMs on host basis. SRM functionality can also be combined with vSphere Replication.
  • Storage Policies – policy-driven VM array-based replication.
  • SRM allows 2 following mechanisms for site replication:
  • Disaster recovery – this mechanism is enabled when your main site goes down all of a sudden. You just switch to your secondary site. Note that the temporary data may be lost.
  • Planned migration – both main and secondary sites should be running. Data won’t be lost.

I hope the image below describes disaster recovery process good enough.

disaster recovery process

You can trigger the replica either manually (whenever you need) or automatically (after the primary site failure). The downtime should be as little as zero, but DON’T EXPECT IT TO BE ABSOLUTELY ZEROED OUT!

To have a disaster recovery plan deployed right, you need to set up these 3 components:

  • vSphere Replication Appliance – the solution for VM asynchronous replication. It is distributed as OVF template. You need to deploy and set up the solution on both sites.
  • Site Recovery Manager – software providing full control over replication. It is deployed on each vSphere site.
  • Storage Replication Adapter – software to ensure Site Recovery Manager interaction with storage arrays. It is supplied as a hardware-vendor-specific driver.

Now, let’s take a closer look at installation and configuration of all those components.

vSphere Replication Appliance configuration

You can install vSphere Replication Appliance from the image. Note that you need to register first at the official VMware website. Today, I won’t bother you with boring installation process. I’d rather shed light only on the key moments.

1. Deploy the OVF template. For that purpose, you need to mount the ISO image that contains that file.

Deploy the OVF template

Apart from the OVF template, ISO contains a bunch of other files. Keep in mind that you need only *.ovf and *.vmdk files.


2. Next, enter VM name and select the destination host and directory.

enter VM name

3. Set the password and enter NTP server IP afterward. Be careful with NTP server address to avoid any troubles with security certificates in future while connecting the host to vCenter Server. If IPs are distributed automatically, you are done with OVF deployment. Otherwise, you need to do some network configuration work.

Set the password

4. Power on the VM now and connect to it. Find out its IP and port number used for connecting it to the console.

Power on the VM

5. Open the web browser and log into vSphere Replication appliance console using the credentials you entered during the *.ovf file deployment.

log into vSphere Replication appliance

6. Enter the credentials to connect vSphere Replication Appliance to vCenter Server.

Enter the credentials

7. Then, confirm the SSL certificate by pressing Accept. You should better always add a certificate signed by trusted authority. I used one that was generated by default.


8. Connect to vCenter Server HTML console. For that purpose, navigate to Home -> Site Recovery and click OPEN Site Recovery.

OPEN Site Recovery

Now, let’s move to Site Recovery configuration.

Setting up Site Recovery Manager

1. Download the image first.

To install it, you need a Windows Server 2016 VM. You should better use that VM only for SRM installation since any other software may mess up Site Recovery Manager performance fairly easy. Next, save the VM disk locally and you are good to start installing that agent.

2. Installation wizard provides you with the link on documentation which you can read while installing SRM. Please, don’t ignore that link. Take your time and read that document carefully in order to avoid any possible inconveniences.


3. For better vCenter Site Recovery Manager performance, you should use a +20 GB eager zeroed thick virtual disk. Today I used a 40 GB device.

The software consumes 1.8 GB of disk space at the beginning, but it will take more disk capacity.


4. Enter Platform Service Controller address.


5. Accept the certificate afterward.


6. Select vCenter Server address located on this Platform Services Controller. In my case, there was only one IP.

vCenter Server address

7. Specify the site address (by default, it matches vCenter Server one) and administrator e-mail. Also, you should add vCenter Server host address.

Specify the site address

8. Enter the SRM Plug-in ID required for a recovery pair creation. Here, I use a default one, but you can also enter a custom ID too.

Enter the SRM Plug-in ID

9. Choose a server certificate type for Site Recovery Manager afterward.

certificate type

10. Select a database server. Today, I use an embedded one.

database server

11. Specify the new database credentials.

database credentials

If you want to run the Service on any other account, you can specify its credentials.

specify its credentials

12. Go to vCenter Server HTML console and verify the Site Recovery Manager instance availability.

 Site Recovery Manager

If everything is ok, you are done with SRM configuration. Let’s configure Storage Replication Adapter (SRA) now.

Configuring Storage Replication Adapter

Storage Replication Adapter is a storage driver that ensures SMR and Storage Virtual Machine interaction. SRA version is hardware-vendor specific (I use Dell PowerEdge R720 today). Find some documentation on adapters here.

To install SRA, enter the VM with the recently installed SRM. Extract the archive and initiate the installation process.

Installation process is pretty straightforward. Just move through all installation wizard steps!


After installation you can review driver description. It contains some prerequisites to ensure SRA smooth operation, so it is nice to look through it.



I guess that it is pretty much it for SRM deployment. Today, I did all boring job, and now I can move to recovery site creation. I discuss how this can be done in this article:

Hey! Found Kevin’s article helpful? Looking to deploy a new, easy-to-manage, and cost-effective hyperconverged infrastructure?
Alex Bykovskyi
Alex Bykovskyi StarWind Virtual HCI Appliance Product Manager
Well, we can help you with this one! Creating a hyperconverged environment is a breeze with StarWind Virtual HCI Appliance (VHCA). It’s a complete hyperconverged infrastructure solution that combines hypervisor (vSphere, Hyper-V, Proxmox, or our custom version of KVM), software-defined storage (StarWind VSAN), and streamlined management tools. Interested in diving deeper into VHCA’s capabilities and features? Secure your personalized StarWind Virtual HCI Appliance demo today!