eG Help

The requests to every application executing on a Siebel server are typically handled by one/more object managers. If the object manager required by an application is not running, then the Siebel server will be forced to reject all requests for that application, causing the end-user to suffer. The SiebelObjMgrTest monitors each of the object managers to ascertain their current state and load.

Measurement	Description	Measurement Unit	Interpretation
Run_state	The current state of this Siebel Object Manager	Boolean	The value 0 for this measure indicates that the object manager is unavailable. While 1 indicates that the object manager is online (i.e., it is available, but not currently running any tasks), 2 indicates that the object manager is running (i.e., it is available and is currently running one/more tasks).
Max_tasks_reached	Indicates whether this object manager has reached its 'maximum tasks' limit or not.	Boolean	This measure is a true indicator of load on the object manager. As long as the value of this measure is 0, it is an indication of an optimal number of tasks currently executing on the object manager. If the value becomes 1, it implies that the 'maximum tasks' limit has been reached. When this happens, eG Enterprise triggers an alarm indicating an overload on the object manager. During such circumstances, the object manager will run out of threads to execute any more tasks, and will therefore be unable to handle subsequent requests.
Num_max_MTS_svr	An MTServer is a multi-threaded component process; this measure indicates the maximum number of MTServers per component per server.	Number
Num_actv_MTS_svr	Indicates the currently active MTServers on this object manager	Number	The value of this should be close to the value of the Num_max_mts_svr measure.
MTS_usage_factor	Indicates the percentage of maximum MTServers that are being actively used by this object manager	Percent	Ideally, the value of this measure should be between 90-100%. A far less value indicates that the object manager is grossly under-utilizing the MTServers. This happens when the object manager does not have enough tasks to run, and is more or less idle.