Configuration Rules for a Mixed HP 9000 / Integrity Serviceguard Cluster, March 2007

• Node C fails.
– No failover occurs; node A takes over the application server load for all users.
– Optionally, the application server instance from node C fails over to node A if supported and
more preferred by the application.
Figure 5. New three-node mixed HP 9000 / Integrity cluster
Node B
Node A
App Server
Database Server
Failover
IPF
App Server *
PA
PA
Node C
OK
Adding two HP Integrity servers to a two-node HP 9000 cluster
This example shows how mixed clusters can help customers to transition from HP 9000 to Integrity
servers. Two new Integrity servers are added to a two-node homogeneous HP 9000 cluster to
transition the database tier to HP Integrity servers while providing high availability for the application
and database tier, each within a set of homogeneous nodes.
Original two-node HP 9000 cluster
Before the HP Integrity nodes are added to the homogeneous HP 9000 cluster, each original cluster
node hosts one critical application. The database and application servers are configured to fail over
to the other node in case the primary node goes down. The failover scenarios are:
• Node A fails.
– Node B runs both database and application servers.
• Node B fails.
Node A runs both database and application servers.
NOT
OK
Node C as
a failover
database
server
OK
19