HP StoreOnce B6200 Maintenance and Service Guide (February, 2014)

It is very important when carrying out service and maintenance activities on a node or storage
controller, that autonomic failover is disabled until the activity is completed. Also, if the maintenance
activity is on the server that is running the active Management Console, that node must be failed
over in a controlled manner before you start.
A controlled failover allows the system to complete a graceful shutdown of all the virtual devices
running on that node, prior to the node being removed from its use in a cluster. The virtual devices
on a failing node are able to cause known error or check conditions to be reported to the backup
application via the appropriate interface. When the service set from the “failed” node is restarted
on the paired node, its power on checks will be quicker and simpler since it was stopped in a
controlled manner.
NOTE: Failover occurs only between nodes in a couplet and not between couplets in a cluster.
Power off
There are no commands within the GUI to control the power off or reboot of a node or cluster. All
the commands should be run from the StoreOnce CLI.
Various commands exist to shutdown a node or a cluster. Please use the correct command to
perform the task you require. More details can be found in the HP StoreOnce Backup System CLI
Reference Guide.
• system shutdown: This gracefully shuts down the cluster by shutting down all services and
StoreOnce service sets before powering down the servers. This means when you power back
up, the deduplication stores will start up from a clean state. Servers will be in the off state,
and will require powering on via the power button or StoreOnce CLI. This command is
appropriate for maintenance tasks that require you to shut down all couplets within a cluster.
In reality, it is only likely to be used when maintaining a single-couplet configuration.
• hardware poweroff nodeX: This uses the iLO port to turn off the server, which does not
allow StoreOnce service sets to shutdown gracefully. This results in deduplication stores shutting
down in an unclean state and requiring integrity checking (an automatic process) on the next
power up. To avoid this condition, the service set on a given node should be failed over to
the backup node using the hardware failover nodeX command, prior to powering down
the node. This command is most likely to be used for maintenance activities because it is node
specific and allows other nodes and couplets in a cluster to remain online.
See also Putting a node into maintenance mode (page 20) for information about putting nodes
and couplets into maintenance mode and making them active again once repairs have been made.
Power on and reboot
There are no commands within the GUI to control the power on or reboot of a node or cluster. All
the commands should be run from the StoreOnce CLI.
IMPORTANT: It is best to ensure that failover is enabled during the power on or reboot sequence
in case of unexpected errors with one of the nodes.
Power on sequence
If the whole system has been powered off and disconnected, the recommended order for connecting
to the PDU is:
1. Network switches
2. Any expansion shelves connected
3. Storage controllers
4. Servers
46 General reference