Data Replication Manager Windows-Based Server with Single Host Bus Adapter - Application Notes

The Single FC HBA Solution Configuration
Application Notes – Data Replication Manager Windows-Based Server with Single Host Bus Adapter 5
The single FC HBA configuration contains the following SPOF:
• The FC HBA within the server
• The fiber optic cable running between the server and the FC switch
• The gigabit interface converter (GBIC), located within the FC switch
• The FC switch on the local site
• The FC switch on the remote site
• The intersite link (ISL)
Some of these SPOF can be mitigated, even with single FC HBA servers. Please refer to the section
titled "No Single Point of Failure (NSPOF) Configurations."
NOTE: Because the HSG80 controllers in a DRM environment require multibus failover, Secure Path
for Windows must be installed on the servers.
Configuration
Software and Firmware
A list of the most current software, firmware, patches, drivers, and so on, for Microsoft Windows and
Microsoft NT in your DRM solution can be found at the following website:
http://www.compaq.com/products/sanworks/drm/index.html
Follow the Software Support Matrices link, then choose your operating system.
Data Replication Manager Configuration
There are no special configuration requirements to allow a single FC HBA server to attach to an
HSG80 controller in a DRM environment. However, keep the following best practices in mind during
the actual configuration:
• Because there is only one FC fabric, each FC HBA establishes two connections within the HSG80
controller pair—one connection to port 1 of the top controller, and the other connection to port 1 of
the bottom controller. Both connections must be on the controller’s unit access list.
• If multiple remote copy sets (RCS) are part of one association set, and if the association set has a
log disk, you must configure the association set for ORDER_ALL. Failure to apply this setting
may lead to data inconsistency at the remote site in the event of rolling disasters. A rolling disaster
is an event that may cause write operations at the initiator to be written out of order at the target
site.
• If multiple RCS are part of one association set, and if one or more RCS is configured with
ERROR_MODE=FAILSAFE, the association set must be configured for FAIL_ALL. This ensures
that all remote copy sets enter the failsafe_locked state at the same time and maintain data
integrity.
Refer to the Compaq StorageWorks Heterogeneous Open SAN Design Reference Guide for
configuration restrictions.