![]() |
|||
![]() |
![]() ![]() |
![]() |
![]() ![]() |
![]() |
![]() ![]() |
![]() |
| ||
Chapter 12Watchdog TimerFor information about how the Watchdog Timer guards against operating system hang, see the following sections: The nhwdt DaemonLow-level system monitoring in the Foundation Services is provided by the Watchdog Timer service. This service is implemented by the nhwdtd daemon. The Watchdog Timer service monitors the hardware watchdog on the nodes of the cluster, at the lights-off management (LOM) level. The nhwdtd daemon monitors a node for operating system hang, but does not monitor the boot process. Hardware watchdogs that operate at the OpenBoot PROM (OBP) level are monitored by the platform's software, not by the nhwdtd daemon. For information about which hardware is monitored at the OBP level and the LOM level, see "Types of Hardware Watchdogs" in the Netra High Availability Suite Foundation Services 2.1 6/03 Hardware Guide. The Watchdog Timer can be configured differently on each peer node depending on your requirements. For information about how to configure the Watchdog Timer, see the nhfs.conf(4) man page. Monitoring by the Daemon MonitorThe Watchdog Timer is monitored by the Daemon Monitor. If the nhwdtd daemon fails, it is relaunched three times. If the nhwdtd daemon fails a third time, the node is rebooted. The nhwdtd daemon operates at a lower priority than the nhpmd daemon, but at a higher priority than the other Foundation Services daemons. | ||
| ||
![]() |