Configuring Alert Rules for Connected Devices

Create custom alert rules to proactively monitor the performance of your Connected Devices fleet. This guide provides step-by-step instructions on how to define specific thresholds for key performance metrics such as latency, packet loss, and jitter, so you are notified the moment an issue occurs.

Overview of Creating a New Alert Rule for Connected Devices

Creating a new alert rule for Connected Devices involves the following steps:

1. Creating and Naming a Rule

2. Defining the Alert Scope

3. Setting Alert Conditions

4. Assigning Severity

5. Saving Your Alert Rule

Creating and Naming a Connected Devices Alert Rule

To begin creating a new alert rule for Connected Devices:

1. Navigate to Manage > Alert Rules and select the Connected Devices tab.

2. In the upper-right corner, click the + Add New Alert Rule button. The configuration panel will open.

3. In the Alert Rule Name field, enter a descriptive name. A good name helps you quickly identify the rule's purpose later, such as High Latency - West Coast Fleet.

4. From the Tests dropdown menu, select one or more tests that this alert rule will apply to.

Defining the Alert Scope for a Connected Devices Alert Rule

The Scope section determines which of your Connected Devices will be evaluated by this rule. You can choose from two options:

• All agents: Select this option for the rule to evaluate data from every agent running the selected test(s). This is useful for creating broad rules that monitor the overall health of your entire device fleet.

• Specific agents: Select this option to limit the rule to a specific set of devices. You can select up to 500 agent IDs. This is useful for targeted monitoring, such as focusing on a single large customer, a new group of deployed devices, or a set of devices in a known problem area.

Setting Alert Conditions for a Connected Devices Alert Rule

In the Alert Conditions section, you define the specific logic that determines when an alert will trigger.

Choosing Metrics and Thresholds for a Connected Devices Alert Rule

You can create one or more conditions based on three key performance metrics: Latency, Packet Loss, or Jitter. For each condition you add:

1. Select a metric from the dropdown menu.

2. Choose an operator, such as > (greater than) or >= (greater than or equal to).

3. Set a threshold value (such as 100 ms or 5 %).

To create more sophisticated rules, such as alerting on a combination of high latency and packet loss, click the + Add Condition button to add multiple conditions.

Combining Multiple Conditions for a Connected Devices Alert Rule

If you add more than one condition, you must specify how they are evaluated together:

• All (AND): An alert triggers only when every condition is met in the same test round.

• Any (OR): An alert triggers when at least one of the conditions is met.

Understanding the Rolling Time Window for a Connected Devices Alert Rule

The rolling time window makes your alerts more robust by requiring a problem to be persistent, which helps reduce noise from transient, one-off issues. This logic is defined by two settings that work together: how many devices must be affected, and how persistent the problem must be before an alert is triggered.

How Many Devices Must Be Affected by a Connected Devices Alert?

This setting determines if you are looking for a widespread issue or a problem with a single device.

• any of [X] agents: Use this to detect widespread issues affecting multiple devices.

• the same agent: Use this to pinpoint a single, consistently problematic device.

How Persistent Must the Connected Devices Alert Issue Be?

This setting defines the observation window and ensures you are only alerted for persistent problems.

• [X] times in [Y] test intervals: An alert will trigger if the condition is met X number of times within the last Y test intervals.

Example: Monitoring an ISP Fleet with Connected Devices Alerts

An ISP monitoring thousands of customer devices could create two separate rules to detect both individual faulty devices and larger regional outages.

• To find a single faulty device: Create a rule where the same agent must have Packet Loss > 5% for 3 times in 5 test intervals. This ensures you are only alerted if one specific device is consistently failing.

• To identify a regional outage: Create a second rule where any of 20 agents must have Latency > 150 ms for 2 times in 3 test intervals. This rule triggers when a significant number of different devices start experiencing high latency.

Assigning Severity to a Connected Devices Alert Rule

Severity levels categorize alerts based on their potential business impact, allowing your teams to prioritize their response. Select the severity level that best corresponds to the alert condition:

• Critical: For service-impacting outages that require immediate investigation.

• Major: For significant performance degradations that could become critical.

• Minor: For less urgent performance issues or early warnings.

• Info: For informational events, often used for tracking and reporting.

For more information, see Alert Rule Severity.

Saving Your Connected Devices Alert Rule

After you have configured all settings, click the Create New Alert Rule button to save it. The rule will become active and will begin evaluating data from the assigned tests on the next test round.

Editing, Duplicating, and Deleting Connected Devices Alert Rules

Alert configuration is often an iterative process. It is a common and recommended practice to review and tune your alert rules over time. To manage an existing rule, navigate to the Manage > Alert Rules > Connected Devices tab and click on the rule's name in the list. From the editing panel, you can modify its settings, duplicate the rule to create a similar one, or delete it.