The following settings are available:

• App.StartStopTimeout

  When you click the Start Application or Stop Application links in the Symantec High Availability view, Symantec ApplicationHA initiates an orderly start or stop of the application and its dependent components. This option defines the number of seconds Symantec ApplicationHA must wait for the application to start or stop. If the application does not respond in the stipulated time, an error is displayed in the Symantec High Availability view.

  A delay in the application response does not indicate that the application or its dependent component has faulted. Parameters such as workload, system performance, and network bandwidth may affect the application response. Symantec ApplicationHA continues to wait for the application response even after the timeout interval is over. If the application fails to start or stop, ApplicationHA takes the necessary action depending on the other configuration settings.

  AppStartStopTimeout value can vary between 0 and 600. The default is 30 seconds.

• App.RestartAttempts

  This option defines the number of times Symantec ApplicationHA should try to restart a failed application or its dependent component. If an application fails to start in the specified number of attempts, Symantec ApplicationHA stops the application heartbeat and communicates the fault to VCS.

  AppRestartAttempts value can vary between 0 and 5. The default is 1.

• App.ShutdownGraceTime

  This option defines the number of seconds Symantec ApplicationHA should wait before communicating the application fault to VCS.

  If a configured application or its dependent component fails, Symantec ApplicationHA tries to restart the component for the configured number of times. If the component fails to start, Symantec ApplicationHA stops the application heartbeat and communicates the fault to VCS. VCS may then restart the managed LPAR depending on the configuration settings.

  An abrupt shutdown may affect the other healthy application components running on the machine. If those components require more time to stop, Symantec ApplicationHA may not be able to stop them gracefully in time before the reboot is initiated. For such cases, you can use AppShutdownGraceTime to delay the managed LPAR reboot so that Symantec ApplicationHA stops all the application components gracefully.

  When an application fails to start, Symantec ApplicationHA initiates a graceful shutdown of all the healthy applications being monitored on the managed LPAR and waits for time specified in this option. A managed LPAR reboot takes place only after all the application components are shut down gracefully or at the end of the grace time, whichever is earlier.

  This setting is applicable to the heartbeat service group that is created when you configure application monitoring using the Symantec ApplicationHA Configuration Wizard. Internally, it sets the DelayBeforeAppFault attribute of the Heartbeat agent resource (VCSAppMonHBRes) in the configuration.

  AppShutDownGraceTime value can vary between 0 and 600. The default is 300 seconds.

• VM.GracefulRebootPolicy

  Internally, it sets the VMGracefulRebootPolicy attribute of the Heartbeat agent resource (VCSAppMonHBRes) in the configuration.

  Use this option to enable or disable ApplicationHA-initiated managed LPAR restart policy. This option defines whether or not ApplicationHA restarts the managed LPAR in response to application and component failures. When a configured application or component fails, ApplicationHA attempts to restart the failed components. If the component fails to start, ApplicationHA then takes the next corrective action.

  If this policy is disabled, and an application or component fails, then ApplicationHA stops sending the heartbeat to VCS. As a result VCS can then restart the managed LPAR.

  If this policy is enabled, ApplicationHA itself invokes a native operating system command to restart the managed LPAR.

  VM.GracefulRebootPolicy value can be Enabled (1) or Disabled (0). The default value is Disabled.

  ApplicationHA uses the following command to initiate a soft reboot:

  # /usr/sbin/reboot

• VM.GracefulRebootAttempts

  This option defines the number of times ApplicationHA attempts to restart the managed LPAR gracefully if the configured application or component becomes unresponsive. The number of restart attempts is time bound and is defined by the option VM.GracefulRebootTimeSpan. The restart attempts count is reset after the reboot time span elapses.

  For example, if the reboot attempts value is 4, the time span value is 1 hour, and ApplicationHA has restarted the managed LPAR once, then the restart attempt count is 3 (initial set value of 4 minus one reboot) for the remaining period of the 1-hour interval. The restart attempts count is reset to 4 at the beginning of the next 1-hour span.

  If the restart attempts are exhausted and the application or component fails within the reboot time span again, ApplicationHA stops the application heartbeat and communicates the fault to VCS. Depending on the configuration, VCS may then restart or fail over the managed LPAR.

  VM.GracefulRebootAttempts value can vary between 1 and 10. The default value is 1.

• VM.GracefulRebootTimeSpan

  Internally, it sets the VMGracefulRebootTimeSpan attribute of the Heartbeat agent resource (VCSAppMonHBRes) in the configuration.This option defines the time interval, in hours, during which ApplicationHA can gracefully restart the managed LPAR for the number of times defined by the option VM.GracefulRebootAttempts.

  VM.GracefulRebootTimeSpan value can vary between 1 and 24. The default value is 1 hour.