Monitoring Your System

Open SDN Controller provides three pages that allow you to monitor the health and performance of your system: the Logs Dashboard page, the Metrics Dashboard page, and the Services Status page. The following topics describe these pages in more detail:

Viewing the Logs Dashboard

From the Logs Dashboard, you can view information for the events that have taken place in your system. To open the Logs Dashboard, select Logs from the main toolbar's Monitoring menu.

Figure 1. Logs Dashboard

Logs Dashboard Components

The following table describes the components that make up the Logs Dashboard.

Component Description

Toolbar

From here, you can:

  • Set the timeframe for which information is displayed in the dashboard

  • Set how often the dashboard's information is automatically refreshed

  • Manually refresh the dashboard's information

  • Revert to the default dashboard layout by clicking the Go to saved default (house) icon

Query field

Allows you to search for event information that contains a particular string. See Running Queries for more information.

Logs widget

Lists the 500 latest events that have taken place in your system. See Viewing Log Events for more information.

Log Summary widget

Indicates the number of events (grouped by severity) that have taken place over the timeframe currently set for the Logs Dashboard. To determine the number of events that are of a specific severity, place your cursor over the corresponding bar in the graph.

Component Summary widget

Indicates the component or device from which events originated and the total number of events that took place on that component or device.

Log Activity widget

Visualizes the number of events that have occurred over the timeframe currently set for the Logs Dashboard. To determine the exact number of events that took place at a certain time, place your cursor over the corresponding bar in the timeline.

Running Queries

By specifying a query, you can view only the event information that contains a particular string. To run a query, enter the appropriate text in the Query field and then click the Search icon or press the Enter key.

Note the following:

  • As you type the string you want to search for, Open SDN Controller suggests additional strings that you can select and search for instead.

  • To search for a string that is part of a longer string, enclose it within asterisks. For example, entering *Closeable* returned the results displayed in the following screenshot. If you had entered Closeable instead, only event information that contained Entries as a separate word would have been returned.

    Figure 2. Sample Query

  • To clear the results of a query you have run, empty the Query field and then click the Search icon or press the Enter key.

  • You can toggle the Query field on and off by clicking the Query button.

Creating Filters

You can create a filter in two Logs Dashboard components: the Logs widget and the Component Summary widget.

From the Logs Widget

    Step 1   Click the table entry for the event you want to base a filter on.

    The Logs table updates, displaying all of the fields available for that event.

    Step 2   Locate the field that contains the value you want to base a filter on.

    For example, say you want to view only events with a severity of 4. In this case, you would need to locate the @fields.Severity table entry.

    Step 3   In the field's Action column, click the Add filter to match this value icon.

    Note that you can create and apply multiple filters to the information displayed in the Logs Dashboard.

    From the Component Summary Widget

      Step 1   Locate the event source you want to base a filter on.
      Step 2   In the Filter By column, click the filter icon to view only the events that originated from that source.

      Setting the Logs Dashboard Timeframe

      Do one of the following to change the timeframe for which information is displayed in the Logs Dashboard:

      • At the top of the dashboard, click the link for the timeframe that is currently displayed in the dashboard. In the resulting drop-down list, select the desired timeframe. If you want to specify a timeframe that is not covered by one of the available options, select Custom, specify the desired timeframe, and then click Apply.


        Note

        From this drop-down list, you can also select Auto-Refresh and specify how often the information displayed by these graphs is automatically refreshed. To manually refresh this information, click the Refresh icon.

      • In the Log Activity widget, click the desired start time. While holding down the mouse, drag the cursor to the desired end time and then release the mouse.

      Viewing Log Events

      From the Logs widget, you can view a listing of the 500 most recent events that have taken place in your system.

      To set which fields are displayed here:

        Step 1   Click the table entry for any event listed in the Logs table.

        The table updates, displaying all of the fields that are available and their current values.

        Step 2   Locate the table entry for the field you want the Logs table to display.
        Step 3   From the Action column, click the Toggle table column icon.

        Repeat these steps to remove a field from the Logs table.

        Viewing Controller Metrics

        From the Metrics Dashboard, you can view graphs that visualize the following performance metrics for the controller, helping you to identify any issues that require attention:

        • CPU usage

        • Memory usage

        • CPU load

        • Heap size

        • Network usage

        • Free disk space

        Figure 3. Metrics Dashboard

        If multiple controller nodes are set up in your system, a separate graph for each of these metrics is displayed for each node.

        To open the Metrics Dashboard, select Metrics from the main toolbar's Monitoring menu.

        Do one of the following to change the timeframe for which information is displayed in the graphs:

        • From the top of the Metrics Dashboard, click Zoom Out. Every time you click this link, the timeframe these graphs cover is expanded.

        • To the right of the Zoom Out link, click the link for the timeframe that is currently displayed in the graphs. From the resulting drop-down list, select the desired timeframe. If you want to specify a timeframe that is not covered by one of the available options, select Custom, specify the desired timeframe, and then click Apply.


          Note

          From this drop-down list, you can also select Auto-Refresh and specify how often the information displayed by these graphs is automatically refreshed. To manually refresh this information, click the Refresh icon.

        • In any of the graphs, click the desired start time. While holding down the mouse, drag the cursor to the desired end time and then release the mouse.

        Viewing Services Status

        From the Services page, you can view the services installed on a controller node, determine whether they are running and, for the services that are running, see how long they have been up. By default, this page is open after you log into Open SDN Controller. To open the Services page when another page is open, select Services from the main toolbar's Monitoring menu.

        Figure 4. Services Page


        Note

        By default, the information displayed on this page is automatically refreshed every 10 seconds.

        Complete the following procedure to determine whether any services are down and need to be restarted.

        1. View the ball icon that precedes a controller node's IP address.

          • If the icon is green, this indicates that all of the services on the node running. You can stop here.

          • If the icon is yellow, this indicates that one or more services are down on the node. Proceed to Step 2.

        2. If necessary, click the node's link to bring up a listing of the four components for which service status is tracked:

          • Metrics framework

          • Logs framework

          • System

          • Controller

        3. Click any component that is preceded by a red ball icon to view a listing of the services installed on that component.

        4. Restart any services that are currently down (indicated by a red ball icon).

        Exporting Diagnostic Information

          Step 1   From the main toolbar's Help menu, select Export Diagnostic Data.

          The zipped TAR file (diagnostic-data.tgz) is downloaded to your default download directory.

          Step 2   Unzip the TAR file to the desired directory.

          The latest diagnostic information for your system is now available for you to view offline.