An event can be defined as a measurable or identifiable occurrence which is relevant to the management of the IT infrastructures and consequently the provision of IT services. Events are typically messages or displays produced by services, configuration items or monitoring tools.

An effective service operation is dependent upon the knowledge of the status of the infrastructures and any variances from the normal service operation. This can be guaranteed by good monitoring and a system of controls which is provided on the basis of two tool types:
 

Click to enlarge

• Active monitoring tools, which interrogate key CIs to ascertain their availability
• Passive monitoring tools, which detect and process alarm signals from configuration items.

A distinction is drawn between the following types of events in the service operation area:

• Events which show that processing has been properly executed (information). These can be completed status reports on a batch processing, registration of a user login or notification that an e-mail has been successfully sent.
• There are also events which flag errors or variances (warning) for example when a user logs on using an incorrect password or in the event of a CPU overload.
• There are also events which display an unusual system response but which is not yet a variance (exception). In these cases the situation needs to be more closely monitored e.g. if processing takes 10% longer than usual.

It is however always important to stipulate clear definitions between normal and critical, as against faulty response by system components