View KPI Performance and Alerts (Custom Rules)

Routing Director generates alerts based on the anomalies in the KPIs in your network. You can view a graphical representation of the performance, and alerts generated for all KPIs associated with a device on the KPI tab (Observability > Health > Custom KPI Collection > Rule Instantiations > Instance-Name > Monitor Instance-Name > KPI (tab)).

You can set triggers and threshold limits for individual KPIs when you define rule parameters. You configure these rule parameters when you define custom rules. To define a custom rule based on your requirements, we recommend that you contact Juniper Networks Professional Services.

You can view the following components on the KPI tab.

• Device Selection and Rule Selection drop-down lists—To view KPI performance and alerts, you must first select a device from the Device Selection drop-down list. After you select a device, you must select a rule (applied on the selected device) from the Rule Selection drop-down list. The rule parameters (or KPIs) associated with the selected rule and device are displayed as accordions on the tab.

• Accordions on the page—The accordions on the page are displayed based on the device and rule that you selected. The name of the accordion is the name of the rule parameters that you set when you defined the custom rule. For more information on the accordions displayed on the page, see KPI Performance Graphs.

• Time Period bar—You can view KPI information for a week, a day, 3 hours, 1 hour, 30 minutes (default), or a custom time period. Use the Time Period bar in the top-right corner of the tab to change the time period settings.

• Refresh icon—Click the refresh icon to update the details on the tab immediately. If you do not use the refresh icon, the page auto-refreshes at an interval depending on the time range for which the information is displayed. The next auto refresh is displayed in in DD/MM/YYYY hh:mm:ss AM/PM format just below the Time Period bar.

The graphs on the KPI Tab display the performance and alerts associated with a KPI. The rule parameters (or KPIs) associated with the selected rule and device are displayed as accordions on this tab. The name of the accordion is the name of the rule parameters that you set when you defined the custom rule.

Each accordion on this tab has three sections.

• The first section displays the name of the accordion, and details of the device and rules.

• The second section displays the alerts generated. Critical alerts are displayed in red and minor in yellow.

• The last section displays a performance graph representing the actual performance of the KPI (associated with the selected device) for the past 30 minutes.

Click any accordion to expand it. The highest severity alert associated with the KPI is displayed under the name of the accordion. The device's keys are listed below the alerts in the order of severity of the events that have occurred on them. You can view up to six keys at a time with the key that is in the most critical state displayed at the top of the list. To view keys that are not listed, click the Select Keys drop-down list and select the interface. However, you must clear a previously selected key to be able to select another key.

Click the toggle button next to the name of an key, to view the performance of that interface in a graph. Depending on the KPI, the graph can either be a line graph or a state graph. You can also zoom into a particular portion of the graph to view more information about events that have occurred. To view details of more than one interface on the graph, click the toggle button next to the name of the interface in the Show Keys list. Details of the interfaces displayed on the graph are also listed in a table below the graph. You can also click the option buttons on the left of an interface's name in the table to highlight the performance for that interface on the graph.

You can view performance and alerts for a week, a day, 3 hours, 1 hour, 30 minutes, or a custom time period. By default, performance for the past 30 minutes is displayed. To change this period, click the Week, Day, 3 hrs, 1 hr, 30 mins, or Custom buttons provided above the graph.

You can view more than 25 data points on a graph related to events (in real time) that have occurred on the interface when you select the 30-minute time period. However, you can only view up to 25 data points related to events when you select a week, a day, 3 hours, 1 hour, or a custom time period (of more than 30 minutes). Data is aggregated to ensure that not more than 25 data points are plotted on the graph at once.

The graph auto-refreshes at an interval depending on the time range for which the information is displayed. See Table 1 for more information. However, you can refresh the graph at any point by clicking the Refresh icon provided above the graph. You can also click the pop-out button next to the graph to open the graph in a new tab. You can view all customizations that you made on the graph in the parent tab, in the new tab.

Alerts, if any, related to the interface is also displayed on the graph, and in the table below the graph. You can also open the graph in a new tab. When you open a graph in a new tab, you can view the following information in the new tab as well:

• Alerts related to the interface.

  Alerts are refreshed across all open tabs simultaneously, when:

  • An alert is refreshed in any one of the open tabs.

  • The last alert fetched was beyond three minutes.

• List of interfaces that you toggled to view from the Select Keys drop-down list.

• Interface that you selected from the table below the graph.

Table 1: Auto-Refresh Rate

Time Range	Auto-Refreshed
Weekly	Every 16.8 hours
Daily	Every 58 minutes
Every three hours	Every 8 minutes
Hourly	Every 3 minutes
Thirty minutes	Every 2 minutes
Custom	No auto-refresh

You can similarly view the graphs and alerts related to the performance of other KPIs associated with the device and rule that you selected.