5.0

Reprotection of MCSC and Fault Tolerant Virtual Machines
To use reprotection with Microsoft Cluster Server (MSCS) and fault tolerant virtual machines, the host
machines on which the virtual machines run must meet certain criteria.
n
You must run a fault tolerant virtual machine and its shadow on two separate ESXi Servers.
n
You can run a cluster of MCSC virtual machines in the following possible configurations.
n
All the virtual machines of the cluster run on a single ESXi Server.
n
If the MCSC cluster is spread over more than one ESXi Server, you can only run one MCSC virtual
machine per ESXi Server, with a maximum of two ESXi Servers in the cluster.
Because of these constraints, reprotection of MCSC and fault tolerant virtual machines is impossible if you do
not enable VMware High Availability (HA) and Distributed Resource Scheduler (DRS). When moving MCSC
and fault tolerant virtual machines across their primary and secondary sites during reprotection, you must
enable HA and DRS, setting the affinity and anti-affinity rules as appropriate.
Reprotection Process
The process of completing reprotection includes two sets of tasks:
n
Reversing the direction of protection groups using the new protected and recovery site configuration.
n
Forcing a synchronization of storage from the new protected site (the original recovery site) to the new
recovery site (to original protected site).
Reversing the Direction of Protection Groups
When a user initiates the reprotection process, SRM instructs the underlying arrays to reverse the direction of
replication. After the replication is reversed, SRM creates placeholder virtual machines at the new recovery
site (the original protected site).
When creating placeholder virtual machines, SRM uses the location of the original production virtual machine
to determine where to create placeholder virtual machines. Placeholder virtual machines are created in a
location that is defined as the placeholder location as part of inventory mapping. If the original production
virtual machines are no longer available (for example because they were deleted by a user), SRM uses the
reverse inventory mappings (from the original recovery site to the original production site) to determine the
resource pools and folders for the placeholder virtual machines. Because of this, these reverse inventory
mappings must be configured prior to running reprotect, otherwise reprotect may fail.
The files for the placeholder virtual machine are placed in the placeholder datastore defined for the original
production site, not the datastore that held the original production virtual machine.
Forcing Data Synchronization
Forcing synchronization of data from the new protection site to the new recovery site ensures that the recovery
site has a current copy of the production virtual machines running at the protection site. Forcing this
synchronization ensures that recovery is immediately possible after the reprotect completes.
Manual Reprotection for vSphere Replication
Reprotection must be manually established for vSphere Replication. After a failover, existing replication is
stopped. Existing replication must be unconfigured and then reconfigured in the opposite direction. It may be
effective to use the original production virtual machine as a copy for physical couriering. In such a case, the
disk names are different, so it is necessary to choose the target for each disk independently. Finally, configure
protection groups and recovery plans in the opposite direction, thereby manually establishing reprotection.
Site Recovery Manager Administration Guide
72 VMware, Inc.