Using the BGP Route Visualization View

ThousandEyes monitors BGP routes from both corporate and service provider networks, giving you visibility into inbound routes to your prefixes and outbound routes to key services and endpoints. ThousandEyes BGP monitoring pulls in data from multiple sources to help identify issues stemming from changes in data routing. Because the BGP route visualization represents how this data traverses the internet, it can assist you in troubleshooting issues that prevent packets from reaching their desired destination.

In addition to creating a standalone BGP test, when you create any test that includes network measurements – including Network, HTTP Server (with network measurements enabled), page load (with network measurements enabled), or DNS server (with network measurements enabled), you get access to the BGP view.

Additional Resources

• For more information about setting up BGP tests, see BGP Tests.

• For an overview of using views, see Getting Started with Views.

• For a comprehensive list of Views 2.0 features, see Introducing Views 2.0.

• For instructions on navigating between layouts, and setting default preferences, see Navigating Between Views Layouts.

• For more information on using multi-service views, see Multi-Service Views.

Overview

The screenshot below shows the BGP layer for a site that is monitored using an HTTP Server test, with network measurements enabled:

1. Title Bar - Includes the title of the test as well as information about when the test was created.

2. Prefix selector – Use this control to change which network prefix is being displayed in the BGP layer. All tests which have network measurements enabled will be shown, including disabled tests. The test selector shows each test listed, with each destination server prefix shown as a selectable child item in the list of tests. In this example, there is only one prefix listed – this corresponds to the network where the target corresponding to the test is located.

   Prefixes Shown:

   • All announced, matching prefixes are shown for each server address. For example, suppose both 1.2.3.0/24 and 1.2.0.0/16 are announced. A connection has one endpoint: 1.2.3.10. Both matching prefixes will be shown in the list of prefixes. Thus, it is always possible to see the routes that will be used if 1.2.3.0/24 is ever withdrawn.

   • All prefixes used by the test over time are listed. If the destination server rotates between different IP addresses or networks, each will be shown in the list. This may be an indication that the test target is hosted on a CDN provider’s servers, such as Akamai.

   • ThousandEyes tests generate data for a prefix for 7 days after the prefix and all related sub-prefixes are removed.

   • If the target IP and prefix changes for a test, the BGP view will keep generating data for 7 days after the change.

3. Timeline - See Timeline.

4. Route Visualization - See Route Visualization.

Timeline

The timeline shows the lesser of 30 days, or the length of time that the test has been enabled, by charting the average values for the metric selected. To change the data displayed, either use the 24h (trailing 24 hours) / 7d (trailing 7 days) / 14d (trailing 14 days) links, click the zoom out and latest buttons, or click and highlight the requested area of the chart, and it will re-draw appropriately. The lower timeline will allow you to zoom in on a particular date and time.

1. Metric Selection - Choose among the various BGP metrics available. These metrics are listed below. Changing the selected metric changes the timeline to reflect that metric:

   • Path Changes: A timeline with a flat set of results is an indication of a very stable network. Spikes indicate an increase in the number of path changes.

   • Reachability: a timeline with a flat set of results (showing 100% reachability) is important for stability. Focus on dips in the timeline to isolate a network event.

   • Updates: A timeline with a flat set of results is an indication of a very stable network. Spikes indicate an increase in the number of updates.

     For more information on metrics that the ThousandEyes platform captures, see ThousandEyes Metrics: What Do Your Results Mean? Changing the selected metric changes the content of the timeline and the coloring of the monitors in the visualization. By default, the Reachability metric is selected.

1. Monitor Overlay - Select specific monitors to show in the timeline view for easier root cause analysis and reporting. For more information about selecting monitors, see the Route Visualization section.

Route Visualization

BGP route visualization illustrates the past and present AS paths taken from all traffic from all monitors travelling to a destination. You can use this visualization to get more context when troubleshooting and identifying routing outages.

1. Monitor Selection – This control is unique to the BGP route visualization. Monitor selection allows you to isolate specific geographies and networks, so you can identify BGP events which occur in relation to a web property you are monitoring.

   By default, the BGP route visualization shows icons from all BGP monitors. You can filter specific monitors using the additional selection menus in the Show field.

2. Grouping – Controls the way the BGP Monitors are shown. They can be displayed as not grouped (grouping by "Monitor") or grouped by the network in which they reside.

3. Highlight - Search for an autonomous system, IP address, prefix or rDNS entry shown in the visualization.

4. Selected Link/Node Details - When either a link or a node is selected in the route visualization, they appear in the link/node details, which is expandable by clicking the down arrow.

5. Monitor and Origin Complexity Controls - This control is also present in the path visualization view, but it effectively allows simplification of the BGP route visualization by collapsing routes between various autonomous systems. The further apart the endpoints on the slider are, the less complex the visualization.

Error States

The path visualization can indicate the effects of a routing change by showing affected routes and nodes. Identifying routing changes can be helpful, for example, in a root cause analysis on impacts on service availability.

Color gradiation is used to indicate how impacted a monitor is by the selected metric. This is determined by the value of the metric. Green indicates a healthy, unimpacted monitor. A red monitor is significantly impacted. Anything in between indicates a partial impact that should be investigated.

Interacting with the BGP Layer

1. Select a test with BGP monitoring. Select BGP from the view options. The default metric shown on the timeline is Path Changes. You can display additional metrics using the Metrics menu.

   
2. The blue bars on the timeline indicate reported path changes. Click on a blue bar to update the BGP route visualization.

   Routes affected by the path change are now red. The affected hop appears as a red dotted line. The affected links are a solid red line.

   

   You can also use the Warnings menu to display any problem areas.

   
3. Click on a hop to expand it. Hover over a monitor to get more information. All items affected by path changes are colored red.

   
4. Select View details of path changes from the details pop-up to view the details of the detected path change.

   
5. Add the affected monitor to the timeline view by clicking on the Monitors button and clicking on the checkbox next to the monitor name. You can also add additional metrics to the timeline to see what else may have been impacted by the detected path change.

   
6. Add another metric to the timeline to further explore the impacts of routing changes.

   

   

Next step: Set up Alerts.

Legend

The following sections describe each symbol used in the BGP route visualization.

Monitors and Nodes

Links and Hops

Alert States

Typically, newly announced routes will be shown in conjunction with withdrawn routes, indicating a routing preference change.