About Site Recovery components (configuration, process, master target)
This article describes the configuration, process, and master target servers used by the Site Recovery service to replicate VMware VMs and physical servers to Azure. This article is applicable to Classic releases.
In modernized architecture, to replicate VMs, you need to create and use an Azure Site Recovery replication server. For information about Azure Site Recovery replication server and its components, see this article.
For disaster recovery of on-premises VMware VMs and physical servers, deploy an on-premises Site Recovery configuration server.
|Components||The configuration server machine runs all on-premises Site Recovery components, which include the configuration server, process server, and master target server.
When you set up the configuration server, all the components are installed automatically.
|Read the configuration server FAQ.|
|Role||The configuration server coordinates communications between on-premises and Azure, and manages data replication.||Learn more about the architecture for VMware and physical server disaster recovery to Azure.|
|VMware requirements||For disaster recovery of on-premises VMware VMs, you must install and run the configuration server as a on-premises, highly available VMware VM.||Learn about the prerequisites.|
|VMware deployment||We recommend that you deploy the configuration server using a downloaded OVA template. This method provides a simply way to set up a configuration server that complies with all requirements and prerequisites.
If for some reason you're unable to deploy a VMware VM using an OVA template, you can set up the configuration server machines manually, as described below for physical machine disaster recovery.
|Deploy with an OVA template.|
|Physical server requirements||For disaster recovery on on-premises physical servers, you deploy the configuration server manually.||Learn about the prerequisites.|
|Physical server deployment||If it can't be installed as a VMware VM, you can install it on a physical server.||Deploy the configuration server manually.|
A process server handles replication data during failover and failback, and installs the Mobility service for on-premises VMware VMs and physical servers.
|Deployment||By default, when the configuration server is deployed the process server is installed.
An on-premises process server is needed for disaster recovery and replication of on-premises VMware VMs and physical servers.
|Role (on-premises)||Receives replication data from machines enabled for replication.
Optimizes replication data with caching, compression, and encryption, and sends it to Azure Storage.
Performs a push installation of the Site Recovery Mobility service on on-premises VMware VMs and physical servers that you want to replicate.
Performs automatic discovery of on-premises machines.
|Role (failback from Azure)||After failover from your on-premises site, you set up a process server in Azure, as an Azure VM, to handle failback to your on-premises location.
The process server in Azure is temporary. The Azure VM can be deleted after failback is done.
|Scaling||For larger deployments, on-premises you can set up additional, scale-out process servers. Additional servers scale out capacity, by handling larger numbers of replicating machines, and larger volumes of replication traffic.
You can move machines between two process servers, in order to load balance replication traffic.
Master target server
The master target server handles replication data during failback from Azure.
- By default, the master target server is installed on the configuration server.
- For large deployments, you can add an additional, separate master target server for failback.