ON THIS PAGE
 

Create the Dashboards and Widgets

Next we'll create dashboards that display live anomalies and down BFD session counts for Fabric and external BGP sessions. We’ll create widgets from the output stages of the State and Match Count analytical processors. The dashboards populate fields automatically, and the widgets use the filters configured at each processor output.

Create Widgets for Fabric-Side BFD Sessions

To create a dashboard widget for a fabric-side BFD session:

  1. From your blueprint, navigate to Analytics > Probes and select your probe from the table. We'll select the BFD Monitoring probe.
  2. From the left pane, select the output stage of Down_Session_Count.
  3. Select the Link !=to_generic filter and click Apply.
  4. Click Create dashboard widget.
  5. Select a dashboard from the drop-down list and click Proceed. You can select one from the list or create your own. For our example, we'll create a new dashboard BFD_Sessions-Monitoring.
  6. From the Create Widget page, assign a Label to the widget. We'll call it Down_BFD_Sessions-Fabric.

    1. Check the Anomalies Only box. Then select the output-stage columns you want visible in the widget.

    2. Under Query, from the Link drop-down list, select the equals (=) filter and enter to-generic . This option displays only fabric-facing interfaces in the dashboard columns.

    3. Click Create to finish creating the widget

    The image below shows the output of the above created widget. Some anomalies are present because certain fabric-facing BFD sessions are down.

  7. From the New Dashboard page enter a Label (name) for the new dashboard. Optionally, you can enter a description. We'll call it BFD-Sessions-Monitoring.
  8. Select the number of columns to display for the widgets. For this dashboard, we're using a Two-column layout.
  9. Click Create Dashboard.
    When finished, the system displays a table displays indicating that several fabric-facing BFD sessions are down, which causes anomalies.
  10. Now, let's create another widget that shows the "total count" of BFD sessions that are down.

    1. From the Processor: Down_Session_Count page, click the Create dashboard widget icon.

    2. Select the existing dashboard BFD-Sessions-Monitoring to add the widget to the layout.

    3. Name the widget. We'll call it Down_BFD_Sessions_Count_Fabric.

    4. Check the Anomalies Only Then select the output-stage columns you want visible in the widget.

    5. Click Create to finish creating the widget. Then click Update Dashboard to add the second widget to the layout.

      Note: Updating the dashboard is required whenever a new widget is added to the layout.
    You successfully created widgets for fabric-side BFD sessions.

Create External BFD Session Widgets

To create an external BFD Session, we'll follow the the same steps as fabric-side BFD sessions widgets. The only difference is the filter configuration. For this configuration, we'll apply the filter link = to_generic at the output stage of both the State and Match Count processors to include only external links in the widget.

View the Dashboard Status

We created dashboard widgets for Down Fabric sessions, Down External BFD sessions and BFD session counts that are down (for both Fabric side sessions and external sessions). The image below displays the dashboard status, showing downtime in all four quadrants.

Exporting and Importing a Service from the Analytics Dashboard (Recommended)

You can export or import a service from the Apstra Analytics dashboard, which contains the probe definitions.

To export a service from the dashboard:

  1. Select your probe from the dashboard table. We selected the probe BFD-Sessions-Monitoring.
  2. Click the Export icon to open the power monitoring dashboard,

    Exporting your dashboard generates a schema with the probe information as shown below.You can import this file to a different instance, if needed.

You finished the necessary steps to create a customer telemetry collection.

Continue to the next section Apstra Rest API Explorer to learn how to easily test REST API endpoints.