Each monitor has individual schedule and timing parameters. Monitors can be configured to have a maintenance period, and are permitted to be unavailable during that time. Each monitor has an alternate schedule that activates when the first alert is started. This provides the ability to more, or less, aggressively test and alert until a resource recovers.

To help determine responsiveness, tests must be completed in a specified number of seconds. This includes the time taken to perform a DNS lookup. Conservative use of DNS names is recommended so that service level responsiveness can be more accurately timed by removing the lookup from the equation.

Monitors count how many sequential failures have taken place. Every time a monitor determines a failure, the count is incremented and checked against the configured "number of failures allowed" before alerting takes place. Once that number has been reached, an alert and/or recovery action will take place. Note, the maximum alert parameter is configured on a per Monitor basis.

In some cases one resource depends upon another to function. For example: many IP resources might depend upon a router. If such an important resource malfunctions, many other resources being monitored would fail, resulting in multiple alerts and/or recovery actions. This single issue should have been localized to the malfunctioning router. By defining a simple hierarchy using a "Group Dependency," malfunctions can be localized to the router and be safely ignored.