HP Insight Recovery 6.3 User Guide

Failover operations
This section explains the difference between planned and unplanned failovers, and provides the
procedure to follow in either case.
Planned failover
A planned failover typically involves an expected outage at the Primary Site. For example, a
planned failover might be necessary to perform scheduled maintenance, or to react to a severe
weather forecast. It can also be used for testing purposes.
A planned failover includes a series of steps performed first at the Primary Site and then the Recovery
Site.
At the original Primary Site:
1. Shut down the applications and operating system on each HP Insight Recovery DR Protected
logical server.
2. From the HP Insight Recovery Sites tab, perform a Change local site to Recovery Site operation
to deactivate and disable DR Protected logical servers. The site role changes to Recovery.
At the original Recovery Site:
1. Ensure that enough resources are available to run the recovery logical servers.
2. From the HP Insight Recovery Sites tab, perform a Change local site to Primary Site operation
to activate DR Protected logical servers. The site role changes to Primary.
Unplanned failover
An unplanned failover typically involves the occurrence of a site-wide event, without prior warning,
at the Primary Site. This event may be a regional disaster (earthquake, massive flood), or a local
problem (power loss or water main leak in the data center).
An unplanned failover includes a series of steps performed first at the Recovery Site and then the
Primary Site.
At the original Recovery Site:
1. Ensure that the DR Protected logical servers at the remote site are no longer running in order
to prevent a split-brain situation.
2. Ensure that enough resources are available to run the recovery logical servers.
3. From the HP Insight Recovery Sites tab, perform a Change local site to Primary Site operation
to activate DR Protected logical servers. The site role changes to Primary.
At the original Primary Site:
After the original Primary Site becomes operational following a disaster or outage:
1. Ensure that the storage associated with each Recovery Group is the destination target for
Storage Replication (use appropriate storage management tools, for example, HP StorageWorks
Command View EVA).
2. From the HP Insight Recovery Sites tab, perform a Change local site to Recovery Site operation
to deactivate and disable DR Protected logical servers. The site role changes to Recovery.
Note that the above sequence assumes that the CMS and managed resources running DR Protected
logical servers survived an unplanned local event (for example, a power loss). If the event is more
severe, resulting in the permanent loss of the CMS or managed resources at the original Primary
Site, reconstruction of the site as a new Recovery Site may be necessary.
If the HP Insight Recovery configuration has been changed since the failover occurred (for example,
a new Recovery Group was created), the sites must be brought into sync by making appropriate
configuration changes at the new Recovery Site. The HP Insight Recovery site configuration export
and import operations can be used for this purpose.
HP Insight Recovery is able to prevent split-brain from occurring during an unplanned failover, by
regulating the auto-power configuration of managed nodes (whether virtual or physical) that are
16 Testing and failover operations