HP Insight Recovery 1.0 Release Notes

Primary Site Logical Server Setup
You must ensure that logical servers selected for DR protection at the Primary Site meet the
following requirements:
• The storage associated with these logical servers must be SAN based.
• If the logical servers you plan to protect with HP IR at the Primary Site have not been
activated, and an OS and applications have not been installed, these steps must be completed
before you can configure these logical servers for DR protection. Once these steps are
completed, deactivate the logical servers.
NOTE: On physical servers that can host DR Protected logical servers, HP recommends that
auto power on is turned off, to prevent split brain. The reason is that while the power is off at
the Primary Site, the remote site may be converted to a Primary Site. If power is restored to the
original Primary Site and auto power on is enabled, the workloads could run at both sites, causing
a 'split-brain' situation.
Storage Setup
In the same way that conflict in the configuration of MAC addresses at the Primary and Recovery
sites is avoided in the preceding “Networking Setup” (page 11) section, conflict must also be
avoided in the configuration of WWNs. The same technique, involving the use of exclusion
ranges, is available for this purpose.
HP IR depends on Continuous Access EVA Disaster Recovery Groups (DR Groups) to enable
failover of logical servers.
In HP IR terminology a Continuous Access EVA DR Group is referred to as a Storage Replication
Group.
To setup HP IR Storage Replication Groups for Continuous Access EVA storage, complete the
following tasks:
1. Boot and data LUNs for VC-based logical servers that are to be DR Protected must be
replicated using Continuous Access EVA.
If a DR Protected logical server at the Primary Site is VC-based, the replicated boot and data
LUNs on the array at the Recovery Site must be presented to the corresponding recovery
logical server.
2. Boot and datastore LUNs for VM-based logical servers that are to be DR Protected, must be
replicated using Continuous Access EVA.
If a DR Protected logical server at the Primary Site is VM-based, the replicated datastore
LUNs on the array at the Recovery Site must be presented to the ESX host at the Recovery
Site that is targeted to run the recovery logical server (ESX guest).
NOTE: Note: LUNs and datastores allocated to a DR Protected logical server are dedicated
to that logical server, and cannot be shared with other logical servers.
NOTE: If a logical server has data disks in addition to the boot disk, then the data disks
and the boot disk must be presented to the logical server, or ESX host, and configured as
members of the DR Group.
3. Create one DR Group for each logical server protected by HP IR.
4. Record the following details about the storage data disk(s) configuration - this information
will be required when you configure Insight Recovery for replicated storage:
• Primary and Recovery Site storage array WWNs
• Command View EVA FQDN names and credentials for the Primary and Recovery Sites
• DR Group name given for the boot and data (or datastore) LUNs of the logical server.
12