HP StorageWorks Data Replication Manager for Sun Solaris 7 and 8 with VERITAS Volume Manager 3.X Application Notes (AA-RQ6AD-TE, March 2004)

Introduction
4 Data Replication Manager for Sun Solaris 7 and 8 with VERITAS Volume Manager 3.xApplication Notes
Introduction
Many Sun Solaris storage area network (SAN) administrators use VERITAS Volume Manager
to administer storage on their DRM solution. VERITAS Volume Manager groups the storage
into pools of free space, called disk groups, which can then be divided into mountable
volumes.
By creating the volumes, VERITAS Volume Manager affects how storage is presented to the
hosts. This has an impact on DRM configuration and failover/failback procedures.
Specifically, when you execute a failover, failback, or role reversal procedure, you will need to
perform some additional steps to make the volumes visible to the host.
This document first presents information on how your DRM configuration may be affected by
the use of VERITAS Volume Manager. It then provides additional steps needed to import,
deport, mount, unmount, and force import volumes when using Volume Manager. These
additional steps are required in 6 of the 11 failover and failback procedures in the HP
StorageWorks Data Replication Manager HSG80 ACS Version 8.7P Failover/Failback
Procedures Guide.
Finally, this document discusses VERITAS Volume Manager and three methods for
administrating DRM-based storage under the Volume Manager. You will need to use one of
the three administration methods to perform the additional steps to import, deport, mount,
unmount, and force import volumes.
About Data Replication Manager
DRM provides a disaster-tolerant solution through the use of hardware redundancy and data
replication across multiple sites. The sites can be near each other or separated by some
distance.
A single DRM configuration requires two HSG80 Array Controller subsystems—one at the
local or initiator site, and one at the remote or target site. For installations with multiple
initiator subsystems, there must be an equal number of unique target sites, one per pair of
initiator controllers.
A DRM configuration consists of paired sites. The initiator site carries out primary data
processing. Target sites are used for data replication. Data processing occurs at the initiator site
and the data is replicated or mirrored to the target sites. If a significant failure occurs at the
initiator site, data processing can be resumed at the target sites, where the data is intact.
DRM uses the peer-to-peer remote copy function of the HSG80 controller to achieve data
replication. HSG80 controller pairs at the initiator site are connected to their partner HSG80
controller pairs at the target site. Remote copy sets are created from units at the initiator and
target sites. These remote copy sets contain storage devices that are mirrors of each other. As
data is written to a unit at the initiator site, it is mirrored to its remote copy set partner unit at
the target site.
DRM requires hardware redundancy. In the event of single component failure at a site, DRM
fails over to a redundant component at that site to allow continued operations. For example, if
one of the dual-redundant Fibre Channel links between the sites fails, DRM switches the data
to the other fabric.