The SystemEDGE agent needs an additional self monitor to monitor a set of processes to sum up the single process attribute values and track the processes in the group. You can configure to monitor the processes as a group by creating entries in the Process Group Monitor table of the Systems Management Empire MIB (needs extra self monitors).
When you create an entry in the Process Group Monitor table, you set the polling interval, a regular expression for populating a process group, and the monitor severity, and the agent monitors all processes that match the regular expression as a group. The agent populates read-only attributes of the Process Group Monitor table with sum group statistics, such as the combined process size, thread count, and real memory usage. If membership in the process group changes (for example, if the agent detects a new process matching the regular expression or if a process in the group becomes unreachable), the agent sends a group change trap. On UNIX, you can also configure the agent to match processes by user name and group name.
For example, you can monitor a group of important processes and be notified when one of the processes goes down. If a process goes down, you can set self monitors on the read-only attributes in the Process Group Monitor table. For example, you can monitor the total CPU time used by a set of processes and set a threshold that corresponds to a service level agreement associated with the corresponding applications.
Note: For more information, see the chapter “Process Group Monitoring."
|
Copyright © 2013 CA.
All rights reserved.
|
|