Alerts for infrastructure: Add, edit, or view host alert information

With New Relic's infrastructure monitoring, you can create alert conditions directly within the context of what you're currently monitoring with New Relic.

For example, if you've used the filter bar to filter to a specific set of hosts, and you notice a problem, you can create an alert from the infrastructure UI. To create the alert, select a saved filter view, and tailor the alert condition directly from the chart you're viewing. This helps you proactively manage and monitor the alerting system for your environment.

Want to try out alert conditions with our infrastructure agent? Create a New Relic account for free! No credit card required.

Create alert conditions for infrastructure

Alert conditions apply to alert policies. You can select an existing policy or create a new policy with email notifications from our infrastructure UI. If you want to use other types of notification channels, create a new policy from within the alerts UI.

To add an infrastructure alert condition to an alerts policy:

1. Go to one.newrelic.com > All capabilities > Infrastructure. Mouse over a chart you want to alert on, select the ellipses icon, and then select Create alert condition.

2. Type a meaningful condition name.

3. Select the Alert type, or refer to the examples to decide which type to select.

4. Create individual filters, or copy all the filters from the entity filter bar to identify the hosts that you want the alert condition to use.

5. Define the Critical (required) and Warning (optional, if available) thresholds for triggering the alert notification.

6. Optional: To create the condition criteria proactively but not receive alert notifications at this time, turn off the Enabled checkbox option.

7. Select an existing policy for the new condition.

   OR

   Select the option to create a new policy and identify the email for alert notifications.

8. Optional: Add a runbook url.

9. Optional: Set the Close open incidents after time limit to automatically close open incidents after a certain amount of time (this defaults to 24 hours for infrastructure conditions).

10. Select Create.

Other infrastructure alert condition methods

You can also use these other methods to create an infrastructure alert condition:

View host alert events

Anyone included in the policy's notification channels receive alert notifications directly. In addition, anyone with permissions for your New Relic account can view infrastructure alert incidents and individual incidents through the user interface.

1. Go to one.newrelic.com > All capabilities > Infrastructure > Events.
2. To change the hosts or time frame, use the search window, entity filter bar, or time functions.
3. From the Events list, select the alert incident.
4. To view detailed information in alerts about the selected incident, select the link.

Update or delete host alert information

To edit, disable (or re-enable), or delete host alert information:

1. Go to one.newrelic.com > All capabilities > Infrastructure > Settings > Alerts.
2. Optional: Use the search window or Select all checkbox to locate one or more alert conditions.
3. Select any of the available functions to edit, disable, enable, or delete the selected conditions.

Use alerts to monitor your entire infrastructure

Alerts provides a single, coordinated alerting tool across all of your New Relic products. This allows you to manage alert policies and conditions that focus on the metrics for entities that you care about the most, such as Docker containers, JVMs, and more.

Add a description

The use of the Description field is available for these alert condition types:

• NRQL conditions: add a description using the NerdGraph API.
• Infrastructure conditions: add a description using the UI or the REST API.

The text you place in an alert condition's Description field is passed downstream to associated incidents and notifications. A description can be used for several purposes, including:

• Capturing the reason for the alert condition.
• Defining the signal being monitored.
• Defining next steps.
• Add metadata to downstream systems.

You can use template substitution to insert values from the attributes in the associated incident event. The template format is {{attributeName}}. For the attributes you can use when creating a description, see incident event attributes.

One available attribute is the special {{tag.*}} attribute. This attribute prefix is used to access any of the tag values that are included with the target signal, or any of the entity tags that are associated with the target signal.

If there are entity tags associated with your incident, then they can be accessed using the entity tag name. An example of this would be {{tag.aws.awsRegion}}. When entity tags are available to use, you see them included with the incident, and displayed when you view the incident.

This field has a maximum character size of 4,000.

Add or edit a runbook URL

The alert condition creation process includes an option for setting a URL for runbook instructions. This lets you link to information or standard procedures for handling an incident. Before adding or updating the link, make sure you use a valid URL.

To add, update, or delete an alert condition's runbook URL:

1. Select an alert condition, and make changes to the Runbook URL link.
2. Save the condition.

In order to be saved, the URL must be a valid URL.

Incident time limit

The incident time limit allows you to define a time period after which incidents will be force-closed. By default, incident time limit is 24 hours for insfrastructure conditions.

To add or update an alert condition's incident time limit:

1. Select an alert condition, and make changes to the incident time limit.
2. Save the condition.

Alert conditions that generate too-long NRQL queries

Alert conditions created for infrastructure rely on behind-the-scenes NRQL queries, and NRQL queries have a 4096-character limit. This means that if your condition generates a very complex NRQL query that filters on many elements (for example, including many hosts or many tags), it will exceed this limit and display an error message saying that the condition failed.

To solve this problem, reduce the number of elements you are using in your alert condition. For example:

For more information, see Best practices for filtering in infrastructure alerts in New Relic's Support Forum.