Device Connectivity Data and Tests Results

To access the Connectivity accordion, navigate to the Observability > Health > Troubleshoot Devices page. Click a device name to access the Device-Name page. Click the Connectivity (accordion) in the Overview tab.

The connectivity accordion on the Device-Name page displays the health of connections from a device in your network to a remote device. The accordion displays the overall status of the device connections at the top-right corner. The status displays Urgent Action Needed if critical events occur within the last seven days or Healthy if no connection issues are detected. You can view connection-specific details when you expand the accordion. You have the flexibility to configure automated connectivity tests when you plan device onboarding or use the Retest button in the connectivity accordion to run connectivity tests after onboarding devices. You can run the tests on all connections (ports) of devices or select connections on which you want to run the test.

After the tests are complete, you can view the results of these tests as links in the Connectivity accordion on the Device-Name page. Table 2 describes the fields in the Connectivity accordion.

Click the health status links for a connection to view details about the faulty connections on the Connectivity Details Page. On the Connectivity Details page, you can rerun tests for specific or all remote endpoints after you resolve the connectivity issues.

The following list explains terms associated with connectivity tests:

• Metrics—Metrics such as delay, delay variance, HTTP timeout, ping (packet) loss, and round trip time (RTT) enable Paragon Automation to collect quantitative measurements to evaluate the quality of a connection.

• Protocols—Protocols such as HTTP, ping, and DNS are used to measure the metrics in a connectivity test. Ping is used to test connectivity from a device in your network to neighboring devices, edge devices, and known hosts in the cloud provider's network. HTTP and DNS protocols are used to test connectivity to Internet endpoints such as DNS service, HTTP service, or other web services.

• Types of remote endpoints—Types of remote devices to which test agents check connectivity. Remote endpoints can be neighboring devices, edge devices, Internet endpoints (DNS servers or web servers), and devices (external hosts) in the cloud.

• Connection—Connections are unidirectional flows of synthetic traffic from a test agent installed on a device to a test agent on another device, from a test agent to a reflector (BGP peering), or from a test agent to an external host in a public cloud. A connectivity test to a remote endpoint, such as neighboring devices, include multiple connections. Depending on the remote endpoint, each connection uses a protocol (such as ping) to check for select metrics (such as RTT). If a single connection (unidirectional traffic flow) experiences issues, the test fails.

• Test Result—Test Results are shown as timeline graphs of multiple key performance indicators (KPIs)—such as error seconds, response time, and packet loss—that indicate the health of a connection type. The KPIs are calculated based on the metric data collected for delay, delay variance, ping packet loss, round trip time, HTTP/ping response time, and HTTP timeout.

To access the Connectivity accordion, navigate to the Observability > Health > Troubleshoot Devices page. Click a device name to access the Device-Name page. Click the Connectivity (accordion) in the Overview tab. Click any hyperlink in the accordion and you are directed to the Connectivity Details page.

The Connectivity Details page contains the following sections:

• Relevant Events—After completing connectivity tests, Paragon Automation generates Critical and Information events for connectivity tests and bad cables. An Informational event denotes that the test passed and a critical event denotes that the test failed. Click View all Relevant Events to view events triggered for all tests to the device's connections.

• Refresh—Paragon Automation automatically refreshes the data every 10 minutes and displays the time for the upcoming round of connectivity data refresh. Alternatively, click the Refresh icon to refresh the connectivity data for the device connections.

• Show Connections Between—Displays the category of remote endpoint (such as edge devices, neighboring devices, and cloud providers), and the number of devices in each remote endpoint category.

  Enable the toggle button corresponding to a connection type to view the health of the connection in the topology view.

• Connections Between Devices—Displays a topology view of all connections from a device. For a remote endpoint, the topology view shows a single line that represents all connections from a device to multiple remote devices. You can perform the following tasks related to connections:

  • Access details of faulty connections on the topology map—After the connectivity tests are run on the onboarded device, the topology view displays the count of faulty connections. Faulty connections appear as red icons on the lines that indicate the connections. You can hover your cursor over the count icon to obtain details of the faults for a connection type.

  • Run connectivity tests—To re-run connectivity tests, click Retest and select All Connections if you want Paragon Automation to re-run connectivity tests on all the connections. Alternatively, you can select a specific connection (Neighbors , Edges, Internet Endpoints, or Cloud Providers) to which you want to re-run the test from the device.

    After the test is complete, the topology view is automatically updated. In addition, the Connections table below the topology view displays the updated information.

    To view additional details (such as a detailed view of the logs raised for events, errors, protocols used, and so on) of a test, click the details icon that appears when you hover your mouse over the time range of the test.

• Connections—Displays a table with details about the connectivity tests run on the device. Table 2 describes the fields you see in the Connections table.

  • To view the test results, click the connectivity status (ERROR, PASSED, or FAILED) on the Connections table. The Test Results for Device-name to Device-name page appears. The Test Results page shows the KPIs and metrics collected from the test connections. You can view results in timeline graphs. See View Connectivity Test Results for more information.

  • To view details of a connection, select a connection in the table. Click More > Detail. The Source to Remote End Point pane displays test time range, source, remote endpoint, test protocol, test result, and number of logs on the Details tab.

    On the log tab, view log details for the connection such as the start time, end time, log level, and log message.

The Test Results for Device-name-1 To Device-name-2 page displays the KPIs and metrics as a timeline graph. KPIs in tests include error seconds, response time, and (packet) loss (%).

You can access the Test Results for Device-name-1 To Device-name-2 when you click PASSED or FAILED status links on the Connectivity Details page. Expand each KPI to view the measurements that are displayed in the timeline graph. Table 3 displays the test metrics for the DNS protocol, Table 4 displays the test metrics for the HTTP protocol, and Table 5 displays the test metrics for the ping protocol.

You can view the following information on the Test Results page:

• Error Seconds—For connectivity tests, the test results are compiled every 1 second. Every second, Paragon Automation checks if an error occurred for a test metric in a connection. If an error occurred in a connection, the Error Seconds graph displays 100%, else 0% for the test result compilation time. The graph plots the error seconds every time test results are compiled for the default test duration of 60 seconds.

  Figure 1: Error Seconds Graph

• Response time—Displays the maximum, average, and minimum response time for DNS, HTTP, and ping packets in milliseconds. Delay variance or jitter is the variance in the amount of time taken by different packets when they traverse from a sending device to a receiving device. The less delay variance you measure in your network, the less latency you experience. Less latency is desirable in voice-based applications such as teleconferencing.

  Figure 2: Response Time Graph

• Loss—The Percent lost metric measures the percentage of pings that are lost out of the total number of pings sent to a remote device. If the percentage of pings lost exceeds 60, then Paragon Automation generates a neighbor ping test alert.

  Figure 3: Loss Graph