Chat now with support
Chat with Support

Foglight Experience Monitor 5.8.1 - User Guide

Installing and configuring Multi-appliance clusters Configuring the appliance Specifying monitored web traffic Transforming monitored URLs Managing applications Foglight components and the appliance Using the console program Troubleshooting the appliance Appendix: Third party software Monitoring the user experience Customizing reports The alarm system Integrating the appliance SOAP-based web services

Variation

This shows how much the alarm value exceeded the alarm threshold when the alarm occurred.

How does the alarm system work?

Whenever an alarm is defined, the system enables an exception check for all resources defined to trigger the alarm. An exception check then occurs whenever a metric is updated for the resource in question. For example, if an alarm is defined on mean command response time for all servers, then exception checks are performed for any server whenever the appliance calculates a new mean command response time. When a metric is updated with exception checks enabled, the new value for the metric is compared against the threshold defined by the alarm. If this threshold is exceeded, then there may be an exception to report. However, before an alarm is generated, the system performs a throttle check.

The throttle check attempts to reduce redundant alarms and ensure that the administrator is only notified of significant problems. One of the first things the throttle check does is check whether this same alarm has been triggered recently. If so, there is no need to notify the administrator again about the same problem. By default, the administrator is notified once every five minutes of a repeating alarm. For each alarm defined, at most 50 alarms can be generated for a given five minute interval. The administrator can adjust this value when defining the alarm. Other checks are performed to verify that this exception is a recurring problem and needs special attention.

Once the system is sure the administrator needs to be notified, the alarm is logged in the database. These alarms can be viewed at any time through the web console. Optionally, the administrator can also receive an email with the details of the alarm or generate an SNMP trap for a SNMP management system.

Setting an alarm

An alarm should always be created with a specific purpose in mind. For example:

The system ships with a set of default alarms already configured for use. These default alarms cover a wide range of potential problems that typically occur on any site. Before setting a new alarm, the administrator should verify that there is not an existing alarm that may already cover their area of interest.

Before defining alarms

There are several issues you should consider before defining new alarms. If you would like to receive real-time notifications via email, the mail server and email groups need to be configured. For more information on both of these topics, see the Foglight Experience Monitor Installation and Administration Guide.

If you would like to generate SNMP traps when an alarm is triggered, then the SNMP server and SNMP community need to be configured (on the Help > Contents page). See “SNMP server” in the Foglight Experience Monitor Installation and Administration Guide.

Related Documents