Discovery and failure detection settings

Use this page to configure the discovery and failure detection settings for a core group. These settings are used to monitor the health of core group members. The discovery protocol establishes network connectivity between core group members of the core group. The failure detection protocol monitors the established network connections. Both the default discovery protocol and the default failure protocol run at regularly scheduled intervals on all started core group members. To view this console page, click Servers > Core Groups > Core group settings > core_group_name. Then, in the Additional Properties section, click Discovery and failure detection.

Use the default protocol providers

Select this option to use the default Discovery Protocol and default Failure Detection protocol provided with the product.

Discovery period

Specifies, in seconds, the time interval that the default Discovery Protocol waits before it recalculates the set of unconnected core group members, and attempts to open connections to those members. Decreasing this value might improve the detection of core group members. However, decreasing this value will also cause the default discovery protocol to consume additional system resources, which might impact performance.

We can specify a value for this property only if you select the Use the default protocol providers

The default value is 60 seconds.

Heartbeat transmission period

Specifies, in milliseconds, the amount of time that elapses between failure detection heartbeats. The default Failure Detection Protocol sends out a heartbeat once during the specified time limit to determine whether a core group member is active. Increasing the length of time between heartbeats might decrease the use of system resources such as CPU. However, decreasing the length of time between heartbeats might improve the detection of failed core group members.

We can specify a value for this property only if you select the Use the default protocol providers

The default value is 30000 milliseconds.

Heartbeat timeout period

Specifies, in milliseconds, the amount of time that constitutes a heartbeat timeout. This value must be an integer multiple of the value specified for the Heartbeat transmission period property. For example, if set the Heartbeat transmission period property to 30000, the value specified for the Heartbeat timeout period property must be n x 30000, where n is a positive integer.

We can specify a value for this property only if you select the Use the default protocol providers

The default value is 180000 milliseconds.

Use alternative protocol providers

Select this option to use an alternative protocol provider provided with the product instead of the default Discovery Protocol and the default Failure Detection Protocol.

You should select this option only if there is an alternative protocol provider available for the platform on which we are running the product.

(zos) Factory class name

If we select the Use alternative protocol providers option, specify the fully qualified class name of the factory used to create the alternate protocol provider.

The fully qualified class name of the factory used to create the alternate protocol provider that is available for z/OS is com.ibm.ws.xcf.groupservices.LivenessPluginZoSFactory.

If we specify the z/OS Cross-system Coupling Facility (XCF)-based protocol provider as the alternate protocol provider, an XCF group is created that establishes monitoring services for all of the core group members, monitors each group member address space, and notifies the high availability manager whenever a member address space shuts down or terminates.

Related tasks

(zos) Select an alternate protocol provider for a core group