The Host Resources MIB (RFC 2790), defined by the Internet Engineering Task Force (IETF), provides agent-less monitoring for generic host systems. Host computers are independent of the operating system, network services, or any software application. The Host Resources MIB provides the standard for monitoring a range of parameters like CPU, memory, disk, software installed, and processes running.
The Host Resources MIB is available out of the box, and thus provides basic monitoring capability immediately with no need for any agents. It can also run along with the CA NSM agents. The Host Resources MIB is useful for providing basic information on less critical systems. For critical systems, you may want to use the other CA NSM agents because they provide more information and send traps to the Event console. The Host Resources MIB does not support traps but sends status information through poll response. Therefore, any system or resource changes may take longer to be detected than with traditional CA NSM agents.
The MIB monitors the following resources:
System
CPU
Memory
Disk file system
Process
You can view the Host Resources MIB on the Management Command Center, Agent View (abrowser), Node View, and MIB Browser.
The Log Agent monitors log data and files by using pattern matching based on regular expressions.
The Log Agent can be configured to monitor ASCII log files and to facilitate the detection of faults in applications running under the monitored system. Besides the monitoring of single log files the agent offers the monitoring of all files in a subdirectory, and the usage of wildcards in file names. Furthermore, the Log Agent can monitor the Windows Event Log, the UNIX Console, files with single lines and files with ASCII control characters.
The Log Agent has only limited support for binary files. It replaces each null character (0x00) by blank before matching against the pattern. These blanks are shown in any trap resulting from a match.
Note: The Log Agent handles Unicode encoded log files as binary files.
The Script Agent provides a rapid way to extend the monitoring capabilities of CA NSM. It leverages or extends existing business logic to manage areas of production systems that are not covered by other CA NSM monitoring components.
The Script Agent's purpose is to run any number of scripts based on a user-defined schedule and to derive a status from the output or return code of each script. The agent monitors these resources and sends an SNMP trap when a user-defined threshold is exceeded.
The Script Agent uses a configuration set that is loaded at agent startup to specify script instances. Script instances define the path to the script, the execution interval, and the return code or the regular expressions that derive a status of Normal, Warning, or Critical. You can specify any number of script instances in a configuration set. The agent reports the status derived for each script instance to the DSM and ultimately to WorldView.
The Script Agent consists of an Agent Technology agent, DSM policy for discovering and managing the agent, .wvc policy for defining the agent classes and states in WorldView, and an Agent View for you to visualize the status of active script instances. The agent also provides an extension to the Unicenter Configuration Manager so you can centrally deploy and manage the agent policy and scripts.
The agent runs under the control of the local Agent Technology Services and implements the policy defined in the configuration set it loads at startup. It maintains the status of the watcher by creating a watcher for each defined script instance and then parsing the script return code or output and comparing that return code or output against the status policy.
The DSM policy discovers the agent and script instances and uses traps or polls to determine the current state of each instance. The .wvc file populates the scriptAgt class to WorldView. Because the scripts often represent elements of key business logic that are being monitored for health and availability, you can include the class in Business Process Views.
Windows, Linux, and most current UNIX platforms support the Script Agent.
The CA SystemEDGE is a lightweight SNMP (Simple Network Management Protocol) agent capable of retrieving, monitoring, and publishing operating system metrics on a wide variety of platforms. It lets remote management systems access important information about the system’s configuration, status, performance, users, processes, file systems and much more. The agent includes intelligent self-monitoring capabilities that enable reporting and managing of exceptions and that eliminate the need for excessive polling.
CA NSM r11.2 supports CA SystemEDGE 4.3. Starting with r11.2 service pack 1, CA NSM supports CA SystemEDGE 4.3 and 5.0.
Note: For more information, see Inside Systems Monitoring in the CA NSM product.
The UNIX System Agent monitors the following resources:
Windows Management Instrumentation (WMI) is an implementation of the Web-Based-Management-Initiative. This movement provides an object-oriented structure for communication between web-based applications, similar to the structure provided by SNMP. WMI fits easily into the Agent Technology architecture, using classes, objects, and properties when describing object instances.
WMI uses namespaces to specify an object instance. A namespace is a string that uniquely identifies a group of instances and classes. It reflects a logical grouping containing specific classes and instances. For example, an operating system process object could have a namespace as follows:
root\CIMV2
The WMI Agent lets you monitor the content of your Windows Management Instrumentation (WMI) to solve problems in real-time. You can specify what resources to monitor as well as WMI events of interest. By providing this level of customization, you can receive alerts on vital areas of your WMI content without having distracting alarms from non-critical resources.
The Windows Management Instrumentation Agent queries your monitored resources for data. This data is compared with thresholds or applied to policies to determine the statuses of the monitored resources. Based on this comparison the agent sends alerts to agent managers in your enterprise.
The Windows System Agent monitors the following resources:
The z/OS system agent enables you to monitor key resources of your z/OS system and provides status, event, and configuration information. The agent can monitor individual resources as well as the health of an entire system, allowing you to quickly determine the cause of a problem. The z/OS system agent also monitors UNIX System Services (USS) resources.
The z/OS system agent puts you in control by allowing you to determine the warning and critical thresholds for each monitored resource. The agent monitors these resources and, whenever a user-defined threshold is exceeded, sends an SNMP trap.
The z/OS Agent runs in IPv6 environments.
The Agent Technology component of CA NSM lets you closely monitor the critical computing resources in your enterprise by configuring the managed nodes to watch and answer queries regarding their resources. Agent Technology uses agents running on remote devices throughout the enterprise to collect data, such as information on file systems, memory utilization, and database performance. This data is distributed to one or more management servers that can interpret the data and take action based on custom configurations.
The Systems Management portion of this chapter explains briefly how each layer of the architecture works, but focuses on the management layer. It introduces you to the tools that allow you to manage, configure, and visualize your computing enterprise.
|
Copyright © 2010 CA.
All rights reserved.
|
|