Service Processor System Monitoring - SurveillanceSurveillance is a function in which the service processor monitors the system, and thesystem monitors the service processor. This monitoring is accomplished by periodicsamplings calledheartbeats.Surveillance is available during two phases:v System firmware startup (automatic)v Operating system run time (optional)System Firmware SurveillanceSystem firmware surveillance provides the service processor with a means to detectboot failures while the system firmware is running.System firmware surveillance is automatically enabled during system power-on. Itcannot be disabled by the user, and the surveillance interval and surveillance delaycannot be changed by the user.If the service processor detects no heartbeats during system boot (for a set period oftime), it cycles the system power to attempt a reboot. The maximum number of retriesis set from the service processor menus. If the failure condition repeats, the serviceprocessor leaves the machine powered on, logs an error, and displays menus to theuser. If call-out is enabled, the service processor calls to report the failure and displaysthe operating-system surveillance failure code on the operator panel.Operating System SurveillanceThe operating system surveillance provides the service processor with a means todetect hang conditions, as well as hardware or software failures, while the operatingsystem is running. It also provides the operating system with a means to detect serviceprocessor failure caused by the lack of a return heartbeat.Operating system surveillance is enabled by default, allowing the user to run operatingsystems that do not support this service processor option.You can also use service processor menus and AIX service aid to enable or disableoperating system surveillance.For operating system surveillance to work correctly, you must set the followingparameters:v Surveillance enable/disablev Surveillance intervalThe maximum time (in minutes) the service processor will wait between heartbeatsfrom the operating system before reporting a surveillance failure.v Surveillance delayThe maximum time (in minutes) for the service processor will wait for the firstheartbeat from the operating system after the operating system has been started,before reporting a surveillance failure.Chapter 7. Using the Service Processor 209
