Specifications

27 Deploying VMware vCenter Site Recovery Manager with NetApp FAS/V-Series Storage Systems
- DR site VMware license services.
- The DR site has time synchronized to the same source or a source in sync with the primary site.
• All required NetApp volumes are being replicated using SnapMirror to the DR site.
• The SnapMirror operations have been monitored and are up to date with respect to the designed RPO.
• Required capacity exists on the DR NetApp controller. This refers to capacity required to support day-
to-day operations that have been planned for in the DR environment.
• All DR ESX hosts already have iSCSI/FC sessions established with the DR storage arrays (physical
arrays or vFilers) and appropriate igroups created.
• Some method exists to isolate the failed primary network from the DR site. This might be necessary if
the event causing the disaster were temporary or intermittent in nature, such as an extended power
outage. If power should be reapplied to the primary site, restarting systems and services might conflict
with the recovered operations that are now running at the DR site.
• Plans have been made for providing users access to the applications and services at the DR site.
Note: It is important to make sure that a server performing the necessary Active Directory FSMO roles is
available at the DR site. For information regarding transferring roles or seizing these roles, see Microsoft KB
255504 http://support.microsoft.com/kb/255504.
8.2 EXECUTING THE RECOVERY PLAN
A disaster has occurred that might require the environment to fail over to the DR site.
In this example, the two ESX hosts at the primary site, Site1, have been physically powered off. A
connection to the vCenter server at the primary site shows the ESX hosts are not responding and the virtual
machines hosted there indicate disconnected. For the purposes of this technical report, after verifying the
ESX servers and virtual machines were no longer accessible, the primary vCenter server was also powered
down to simulate a complete failure of Site1.
Figure 14 The ESX hosts and VMs at the primary site are down
As mentioned above the location of the placeholder VMs on the ESX hosts determines where they will
startup. If not using DRS the ESX host used to start each VM in the recovery plan is the host on which the
placeholder VM is currently registered. To help make sure that multiple VMs can be started simultaneously
you should evenly distribute the placeholder VMs across ESX hosts before running the recovery plan. SRM
will startup two VMs per ESX host, up to a maximum of 18 at a time, while executing the recovery plan. For
more information about SRM VM startup performance see VMware vCenter Site Recovery Manager 4.0
Performance and Best Practices for Performance at www.vmware.com/files/pdf/VMware-vCenter-SRM-WP-
EN.pdf
1. Before executing the DR plan, the administrative team must take predetermined steps to verify that a
disaster has actually occurred and it is necessary to run the plan. After the disaster is confirmed, the
team can continue with execution of the recovery plan.