HP StoreAll Storage Network Best Practices Guide

Fusion Manager VIF failover
The Fusion Manager uses a VIF to enable it to fail over across the cluster. A single cluster-wide IP
address is chosen for the Fusion Manager. The node that is running the active Fusion Manager
then establishes an active VIF for the Fusion Manager’s IP address. When the Fusion Manager
needs to fail over to a different node, the following occurs:
1. The original node hosting the active Fusion Manager disables its FM VIF.
2. The new node that will host the active Fusion Manager enables its FM VIF.
Example FM failover
The following example illustrates what occurs when the active Fusion Manager migrates between
nodes. The example shows an StoreAll 9730 platform in a flat network topology. The IP addresses
and other identifiers have been chosen for illustration purposes and could be different on a customer
installation.
In this example, the active Fusion Manager has been assigned a cluster-wide IP address of
10.30.214.201. The active Fusion Manager starts on StoreAll node 1 and will migrate to StoreAll
node 2. Figure 2 (page 8) shows the initial state of the VIFs on StoreAll node 1 and StoreAll node
2. The interfaces in bold (bond0) are active on the StoreAll node, while the other interfaces are
the standby interfaces put in place for failover.
Figure 2 FM failover example — start state
In Figure 2 (page 8):
• StoreAll node 1 is servicing file client requests to 10.30.214.202 (StoreAll node 1–bond0:2).
• StoreAll node 2 is servicing file client requests to 10.30.214.203 (StoreAll node 2–bond0:2).
• Fusion Manager is active on StoreAll node 1, accepting requests to 10.30.214.201 (StoreAll
node 1–bond0:0).
Initially, StoreAll node 1 has three active interfaces and one standby interface:
• bond0is the cluster network interface
• bond0:0 is a cluster network VIF for the active Fusion Manager
• bond0:2 is a user network VIF for file services from StoreAll node 1
• bond0:3 is not active, but is provisioned for failover of file services duties from node 2
Initially, StoreAll node 2 has two active interfaces and two standby interfaces:
• bond0 is the cluster network interface
• bond0:0 is not active, but has been provisioned to support Fusion Manager failover
• bond0:2 is a user network VIF for file services from StoreAll node 2
• bond0:3 is not active, but is provisioned for failover of file services duties from StoreAll node
1
The ibrix_nic command displays the cluster-wide configuration of the network. Before failing
over the FM, the output would be similar to the following:
8 Overview of HP StoreAll Storage networking