Event Management uses the RMAPI (Resource Monitor API) to communicate with and control resource monitors (RMs).
Resource monitors are of two types: external and internal.
RSCT ships with seven external resource monitors. They supply data to the EM daemon. The seven external RMs are:
This monitor provides the state of the SP hardware. This information is obtained from the PSSP hardware monitoring subsystem (hardmon). The resource variables are of type state, and sent directly to the Event Management daemon as a message. The process name is hmrmd. It is started by the Event Management daemon. The executable is located in /usr/sbin/rsct/bin/haemRM.
This monitor examines processes that are running a particular application. The resource variables can be used to determine whether or not a particular system daemon is running. The resource variables are of type state, and sent directly to the EM daemon as a message. The process name is harmpd. It is started by the Event Management daemon. The executable is located in /usr/sbin/rsct/bin/haemRM.
This monitor provides switch, IBM Virtual Shared Disk, LoadLeveler, processor on-line information, and internal variables. It uses shared memory, as it only reports on resource variables of type counter and quantity. The process name is harmld. It is started by the Event Management daemon. The executable is located in /usr/sbin/rsct/bin/haemRM.
This monitor supplies the resource variables of the Problem Management subsystem (PMAN). On your behalf, pmanrmd can run a program, script, or command to report on various aspects of the system. The resource variables are of type state, and sent directly to the Event Management daemon as a message. The process name is pmanrmd. It is started by the AIX SRC. The subsystem name is pmanrm.syspar. The Perl script is located in /usr/lpp/ssp/bin/pmanrmd.
This monitor provides resource variables that represent AIX operating system resources. This is a daemon (harmad) with a connection type of server. The SPMI library provides the means to directly query AIX structures (such as the kernel and Logical Volume Manager) to supply data for a range of operating system resource variables. The aixos resource monitor calls the SPMI library (part of the perfagent.tools component). The process name is harmad. It is started by the AIX SRC. The subsystem name is haemaixos.syspar. It is located in /usr/sbin/rsct/bin/haemRM.
This monitor supplies a resource variable that represents the state of CSS error log entries. This is a command-based resource monitor with a connection type of client.
This monitor provides a resource variable that represents the modification state of SDR classes. This is a command-based resource monitor with a connection type of client.
The two internal RMs are:
This monitor supplies the resource variables that correspond to node and network adapter state. The information is obtained directly from Group Services by subscribing to the system groups hostMembership, enMembership, and cssMembership.
This monitor supplies the IBM.PSSP.Response resource variables. The IBM.PSSP.Response.Host.state resource variable is updated based on information coming from the adapter membership information (en0 adapter in particular) supplied by Group Services. The IBM.PSSP.Response.Switch.state resource variable is updated based on the switch_responds class in the SDR. This SDR class is updated by the switch daemon itself.
The Host_Responds daemon (hrd) obtains the Ethernet state information by subscribing to the IBM.PSSP.LANAdapter.state resource variable through the EMAPI. It receives events from the Membership monitor.