Implementation; Device Configuration Preparation - HP StorageWorks 12000 - Virtual Library System EVA Gateway Manual

• The VLS will automatically save, within 1 hour, the current configuration and licenses to a hidden
virtual cartridge stored on the back-end disk arrays.
• In the event of a node0 disaster (node failed and was replaced with a new one, lose of node0
operating system, Quick Restore of new operating system, etc) the node0 recovery process is:
• Replace/repair node0 and set it to "master" ID through console/serial as usual.
• From that point on, the rest of the device recovery process is fully automatic. The device will
automatically restore its previous configuration and previous licenses (which will generate an
extra automatic reboot).
• The resulting repaired device will have the identical configuration as before including serial
numbers, front-end FC WWPNs, virtual device configuration, licenses, etc. When using the
deduplication-enabled replication feature, you must still manually re-enter the SLAVE node
TCP/IP addresses because these are currently not part of the save/restore configuration system
and thus not part of the warm failover).
This process makes it much quicker and easier to replace a failed node0. One warm failover scenario:
purchase a spare unconfigured node (so it must be ordered separately and unracked). Pre-install and
pre-wire this spare node into the VLS but leave it unconfigured and not powered on. In the event of
a node0 failure, manually unplug the external Fibre Channel and LAN connections from the failed
node0 and move them to the replacement node0 and then power it on and configure the replacement
node as the master. The automatic configuration restore takes care of the rest.
NOTE:
Manually transferring the external Fibre Channel connections from the original node0 to the
replacement node0 guarantees that there will be no backup SAN IOP problems because it will
effectively be the same node0 with the same WWPNs plugged into the same customer Fibre Channel
switch ports. If you have the replacement node0 pre-wired with its own external Fibre Channel and
LAN connections (so you do not have to transfer the Fibre Channel connections from original to
replacement node0) this can cause IOP problems because this will not be identical to the original
node0. It will be using different Fibre Channel switch ports, and some operating systems that use
switch-based SAN addressing (e.g., HPUX older than v11.31) will see this as a brand new device
with new device paths; backups configured to the old paths will no longer work.

Implementation

Device Configuration Preparation

Before installing and configuring the VLS device, review the following steps needed to scope the
configuration requirements for the VLS and the backup application following the best practices.
If deduplication and replication will be configured on the VLS, review the
and Replication Preparation
Basic VLS device configuration
The following steps highlight the considerations for designing the required virtual
libraries/drives/cartridges and standard device configuration.
• Identify the number of required virtual libraries. This could be based on the number of separate
SAN fabrics or separate backup application instances/types. See
raries.
for more information.
HP StorageWorks VLS and D2D Solutions Guide
Deduplication Preparation
Single Library vs. Multiple Lib-
115