Information on Client (agent-based) and Agentless Exception Monitoring in Operations Manager 2007 on the System Center WIKI.
- Related Documentation
- Types of Client Monitoring
Related Documentation
Types of Client Monitoring
Client Monitoring comes in 3 varieties in Operations Manager 2007:
- Agentless Exception Monitoring
- Collective Client Monitoring
- Business-Critical Client Monitoring
Agentless Exception Monitoring
This is effectively collection of crash-and-hang data from systems without the use of a SCOM agent. AEM gathers application crashes and hangs from Windows operating systems and the applications running on them. All supported Windows operating systems have a service called Windows Error Reporting. Windows Error Reporting uses the Windows Error Reporting client, also known as the Watson client, to gather information about application and operating system crashes. AEM can then forward the crash information to Microsoft for analysis.
Agent-based Client Monitoring
There are two types of agent-based client monitoring: Collective Client and Business-Critical Client Monitoring.
Collective Client Monitoring
Collective Health Monitoring is performed by gathering event and performance data from many machines and aggregating the data together based on groups of systems for reporting and analysis. In short, Collective Health Monitoring does NOT generate alerts on monitoring data collected
For example, individual memory performance data is gathered from Windows XP and Windows Vista clients on different types of hardware. Collection Health Monitoring will aggregate this data together and provide reports based on memory performance for specific groups of systems, such as by operating system or by hardware vendor. This makes analysis of overall performance easier than digging through long lists or individual system performance reports.
Business-Critical Client Monitoring
In some cases, client systems are critical to the business’ operations. Point-of-sale terminals, kiosks, bank ATMs, or certain employee workstations, such as bank traders or manufacturing engineers, are examples of critical client systems. If a critical client system goes down or has significant performance problems, the business will lose money. These systems need to be monitored as if they were servers. This type of monitoring generates monitoring alerts as if the client systems were server computers.