9.5 HP P4000 Storage Replication Adapter for VMware SRM Deployment Guide (AX696-96101, September 2011)

Volumes recognized by the HP P4000 SRA
Volumes must be configured with scheduled remote snapshots or they will not be considered for
SRM by the SRA. The HP P4000 SRA recognizes the scheduled remote snapshots as replicated
LUNs for SRM to manage. Volumes that are being remote copied only manually are not updated
consistently enough to be useful for disaster recovery and SRM.
If a volume has scheduled remote snapshots, and you create an ad hoc manual remote snapshot
of that volume, then the HP P4000 SRA will recognize the manual remote snapshot along with the
scheduled remote snapshots. Such ad hoc remote snapshots may be useful to test failover with a
more recent copy than the schedule has, or to push the latest updates to the remote site before a
foreseeable disaster occurs.
Volumes ignored by the HP P4000 SRA
Remote snapshots within the same HP P4000 SAN Solution management group are ignored. These
snapshots are not considered disaster recovery copies since they are usually created to clone a
volume in the HP P4000 SAN Solution, and copies that reside on the same SAN are not useful for
disaster recovery.
Testing failover at the recovery site
SRM performs test failovers at the recovery site when requested. When test failovers are requested,
the HP P4000 SRA will perform the steps listed.
1. Select the replicated volumes.
2. Identify the latest complete Remote Copy snapshot.
3. Delete any temporary writable space on that snapshot to ensure an unedited snapshot is
presented to ESX servers.
4. Configure authentication for ESX servers to directly mount snapshots.
5. When testing stops, to conserve space on the SAN, delete the temporary writable space that
was used during the test.
Actual failover at the recovery site
In the event of a disaster, the SRM can perform actual failover when requested to. When actual
failover is requested the HP P4000 SRA will perform the steps listed.
1. Select the replicated volumes.
2. Identify and remove any incomplete remote copies that are in progress and present the most
recent completed Remote Copy as a primary volume.
3. Convert remote volumes into primary volumes and configure authentication for ESX servers to
mount them
If an actual failover does not run completely for any reason, the failover can be called many times
to try to complete the run. If, for example, only one volume failed to restore and that was due to
a normal snapshot being present, the snapshot could be manually deleted and the failover be
requested again.
8 Using the HP P4000 SRA for VMware SRM 1.x/4.x