Managing Serviceguard Seventeenth Edition, First Reprint December 2009

Planning for Expansion
You can add packages to a running cluster. This process is described in “Cluster and
Package Maintenance” (page 297).
When adding packages, be sure not to exceed the value of max_configured_packages as
defined in the cluster configuration file; see “Cluster Configuration Parameters ”
(page 139). You can modify this parameter while the cluster is running if you need to.
Choosing Switching and Failover Behavior
To determine the failover behavior of a failover package (see “Package Types”
(page 68)), you define the policy that governs where Serviceguard will automatically
start up a package that is not running. In addition, you define a failback policy that
determines whether a package will be automatically returned to its primary node when
that is possible.
The following table describes different types of failover behavior and the settings in
the package configuration file that determine each behavior. See “Package Parameter
Explanations” (page 263) for more information.
Table 4-2 Package Failover Behavior
Parameters in Configuration FileSwitching Behavior
• node_fail_fast_enabled set to no. (Default)
• service_fail_fast_enabled set to NO for all services. (Default)
• auto_run set to yes for the package. (Default)
Package switches normally after
detection of service, network, or EMS
failure, or when a configured resource
dependency is not met. Halt script runs
before switch takes place. (Default)
• failover_policy set to min_package_node.
Package fails over to the node with the
fewest active packages.
• failover_policy set to configured_node. (Default)
Package fails over to the node that is
next on the list of nodes. (Default)
• failback_policy set to automatic.
Package is automatically halted and
restarted on its primary node if the
primary node is available and the
package is running on a non-primary
node.
• failback_policy set to manual. (Default)
• failover_policy set to configured_node. (Default)
If desired, package must be manually
returned to its primary node if it is
running on a non-primary node.
166 Planning and Documenting an HA Cluster