Node-to-Node Alerts

  • 3 minute(s) read
Prev Next

The Node-to-node test type supports a wide range of alert options and triggering conditions, which differ from other test types. Follow the steps below to configure custom alerts for your Node-to-Node tests.

Configuring Node-to-Node Alert

  1. Edit the properties of a Node-to-Node test, locate the Alerts section, and turn off inheritance if it is enabled. Then enable alerts and click New Alert.

  2. Select an Alert Type to specify what metric will be checked to generate alerts.
    image.png

    • Test Failure - alert when number of test failures exceeds threshold.
    • Availability - alert when node availability drops below threshold.
    • Ping RTT - alert when ping roundtrip time exceeds threshold on specified number of nodes.
    • Ping Packet Loss - alert when ping packet loss exceeds threshold on specified number of nodes.
    • % Downtime - alert when specified number of nodes experience downtime in excess of threshold.
    • Path # ASNs - alert when the number of Autonomous Systems traversed in Traceroute test exceeds threshold.
    • Path # Hops - alert when the number of hops performed in Traceroute test exceeds threshold.

  3. Specify the Node Type to alert on.

  4. Configure the alert Trigger. The trigger options will vary depending on the alert type selected. In this example we will configure a trigger for a Ping RTT alert.

    • Hop - determines which hop(s) to look at when checking Ping RTT.
      • Any - Ping RTT from any hop may contribute to alert trigger
      • Index - Check Ping RTT for a specific hop by its number
      • Hostname - Check Ping RTT for a specific target host (input exact name or Regex)
      • IP - Check Ping RTT for a specific target IP address
      • Last- Check Ping RTT for the last hop only
    • Choose the type of change that may trigger the alert.
      • Specific Value - Compare RTT from the selected timeframe against specified thresholds.
      • Trailing Value - Compare RTT from the selected timeframe against performance of recent test-runs (e.g. how does performance compare to the average RTT from all test runs in the past 24 hours?) Learn more about Trailing Value alerts
        • Utilize per node historical data - if enabled, performance is only compared against recent performance of the same node. If disabled, performance is compared against recent performance of all nodes in the mesh.
      • Trend Shift - Rather than specifying trigger criteria, Catchpoint's Trend Shift algorithm analyzes historical test performance and triggers alerts when it detects significant change in performance trends.

  5. Configure Conditions.

    • Cell - Alert will trigger if results from any single cell (node pair) exceeds threshold. If selected, then specify the number of occurrences in a single cell that must occur during the selected timeframe to trigger alert.
    • Node - Alert will trigger if results to/from any single Node exceed threshold. If selected, then specify the number of cells for a specific node that must exceed threshold, the number of runs that must experience the issue during the timeframe, and whether these runs must be consecutive.
    • Matrix - Alert will trigger if aggregated results from all nodes in matrix exceed threshold. If selected, specify whether Matrix data must exceed threshold on consecutive runs.

  6. Configure Timeframe The timeframe specifies the sample of test runs we are checking against our alert criteria. (Test runs from the last 5 minutes up to the last 12 hours).

Below is a sample configuration: This alert checks Ping RTT for each Cell in the Matrix on a rolling 30-minute basis and triggers an alert if any two consecutive runs include hops that exceed 5ms (warning level) or 10ms (critical level).

image.png

6. Configuring Additional Options

In addition to the configuration settings above, you can set up the following for Node to Node Alerts:

  • Reminders - select how frequently you want to receive additional Warning and Critical alert messages when a test remains in the original alert-triggering state.
  • Scatterplot - select whether to include a scatterplot of recent test runs in the alert email.
  • Instructions - optional additional text that you can include in the alert email to instruct the recipient on how to proceed. You can specify Custom instructions unique to this alert configuration, or you can use a Template.
  • Notification Settings
    • Recipients - specify recipients for this alert configuration. You can also include the default recipients. If selected you must add one or more recipient groups to receive the different alert messages (warning, critical, & improved.)
    • Default Recipients - send this alert to default recipients only (these are recipients defined in the main Test Properties window under Alerts.)
    • Do not notify, log only - no emails are sent. Alerts can be seen in Dashboards and the Alerts log.