The SystemEDGE agent supports a state management model for self monitors and process monitors fully integrated with the overall CA Virtual Assurance Management Model. The agent aggregates multiple monitors of different severities into a single Managed Object. This object has a state corresponding to the breached monitor with the worst severity.
The agent calculates individual monitor states according to an assigned severity value. The resultant states can be one of the following:
Note: If a monitor has a severity of none, the state toggles between up and down.
The Aggregate table of the Systems Management MIB uses the object class, instance, and attribute values to aggregate monitors with the same values into one entry. This entry represents a monitored object, for which it maintains an aggregate state.
Note: If you do not enter values for the object class, instance, and attribute in a monitor, the agent populates them with meaningful default information. Default self monitor values are based on the monitored OID using a sysedge.oid file that maps a monitored OID to instance, class, and attribute values. Default process monitor values are based on the process regular expression and monitored attribute.
The Aggregate table updates the current state in the table and sends a state change trap only when a threshold breach creates the worst state of all monitors for an object. For example, assume that you are monitoring CPU usage with three monitors; one for 60 percent (assigned a warning severity), one for 80 percent (critical severity), and one for 100 percent (fatal severity); and the agent returns 82 percent CPU usage. This value causes a threshold breach for the 60 percent and 80 percent monitors. However, the agent only sends one state change trap for the 80 percent monitor and changes the aggregate state to critical.
|
Copyright © 2013 CA.
All rights reserved.
|
|