Configure alerting options
There are several ways you can modify the kinds of alerts you receive and when to receive them.
You can configure alerting options via the Service Settings menu:
- Click on the name of the service you want to change alert settings for. You can do this either from the Dashboard or the Services window located on the left-hand side of the Services tab page.
- Click on the Settings sub-tab for your service to load the Service Settings menu.
- Select the menu option you wish to change, as described below.
The following options may be displayed:
- Strikes Before Error: The Monitoring Settings area will have a Strikes Before Error drop-down menu to select the number of strikes before an error is logged and an alert is issued. By default, 3 is selected, which is the recommended number. Depending on your error tolerance, you can select 1, 2, 3, or 4 strikes before error. If '1' is selected, then any time the service encounters an error, you will be alerted. See What_is_the_three_strike_rule for more information.
- This area allows you to configure alert escalation levels to specify the timing with which you want to receive alerts for your services.
- There are four areas to enter alert contacts:
- Diagnostic/Warning Email Contact - these contacts are alerted immediately when any error is detected.
- Alert Escalation Level 1 - these contacts are also alerted immediately when any error is detected.
- Alert Escalation Level 2 - these contacts are alerted if the error continues for a specified number of minutes, ranging between 5 and 360 minutes at varying intervals.
- Alert Escalation Level 3 - these contacts are also alerted if the error continues for a specified number of minutes, usually a longer period than for Level 2.
- Typically, the threshold for Alert Escalation Level 2 should be at least as long as, if not longer than, the monitoring interval.
- For example, if you have a 20-minute monitoring interval and plan to utilize Alert Escalation Level 2, if you set the threshold to 5 minutes, those contacts will be alerted immediately upon any error. Set the value to at least 20 minutes.
- Set the threshold for Alert Escalation Level 3, if you plan to use it, to a value longer than Alert Escalation Level 2.
- You can add or remove contacts as needed to each escalation level by following the instructions in Add_an_alert_contact_to_a_single_service or Add_an_alert_contact_to_multiple_services_simultaneously and Stop_receiving_alerts.
Alert and Notification Settings
- Alerts for specific error types: Under the Notification Settings heading you can specify for the four types of alerts (timeouts, connectivity errors, content errors, or objective/SLA violations) whether you want to send an alert, send a warning, or ignore the error. These are individually configurable. The recommended option is noted in the drop-down menu and is selected by default. This follows the Strikes before error setting.
- Continuity of alerts: All service levels can choose whether to alert continuously or to stop sending alert messages after a certain number have been issued. These options are mutually exclusive. If you elect to stop after X alerts, you can select from 1, 2, 3, 4, or 5 alerts.
- OK messages: All service levels can also choose whether to send an OK message when the errors stop. See Why_do_I_get_an_OK_after_X_minutes for more information about OK messages.
- Localized Agent Errors: Check the Localized Agent Error box to be notified if a particular agent experiences errors X times in a row, where X is the number of strikes before error as set under Monitoring Options.
- Statistical Variation Errors: Check the Statistical Variation Error box to send an alert when the performance exceeds the weekly average by a certain percentage. You can change this percentage using the drop-down menu; the range is 25 to 1000 in varying increments.