Amazon ECS/ECR monitoring integration | New Relic Documentation

Important

Enable the AWS CloudWatch Metric Streams integration to monitor all CloudWatch metrics from your AWS services, including custom namespaces. Individual integrations are no longer our recommended option.

New Relic infrastructure integrations include an integration for reporting your Amazon EC2 Container Service (ECS) data and your Amazon EC2 Container Registry (ECR) data to New Relic products. This document explains the integration's features, how to activate it, and what data can be reported.

Important

New Relic also offers an ECS on-host integration, which reports a different data set than this cloud integration. For complete ECS monitoring, we recommend enabling both integrations.

Features

With New Relic's ECS/ECR monitoring integration, you can monitor reserved vs. utilized capacity, task execution, and registry of containers. AWS integration data is also available for analysis and chart creation in New Relic One.

Activate integration

To enable this integration, follow standard procedures to Connect AWS services to New Relic.

If you have services running on ECS, you can also enable monitoring of those services.

Important

In AWS, you have to opt-in for the new ARN format (announcement) to differentiate services with the same name in different clusters. If not, you could have data collision

Configuration and polling

You can change the polling frequency and filter data using configuration options.

Default polling information for the Amazon ECS/ECR integration:

New Relic polling interval: 5 minutes
Amazon CloudWatch data interval: 1 minute or 5 minutes

Find and use data

To find this integration's data, go to one.newrelic.com > Infrastructure > AWS and select one of the Amazon ECS/ECR integration links.

You can query and explore your data using the ComputeSample event type. Use a provider value of EcsCluster for cluster data, or a provider value of EcsService for cluster-service data.

The integration collects these ECR/ECS definitions:

Name	Description
`ClusterName`	This dimension filters the data you request for all resources in a specified cluster. All Amazon ECS metrics are filtered by `ClusterName`.
`ServiceName`	This dimension filters the data you request for all resources in a specified service within a specified cluster.

For more on how to find and use integration data, see Understand integration data.

Metric data

The ECS/ECR integration collects the following data:

Tip

For full descriptions of these metrics, see Amazon's documentation on ECS and ECR.

Cluster and service metrics

Name	Data type
`CPUUtilization`	percent
`MemoryUtilization`	percent

Cluster metrics

Name	Data type
`CPUReservation`	percent
`MemoryReservation`	percent

Service metrics

Name	Description
`Active Service`	The number of services that are running on the cluster in an `ACTIVE` state
`Pending Tasks`	Number of tasks in the cluster that are in `PENDING` state
`Running Tasks`	Number of tasks in the cluster that are in `RUNNING` state
`Registered Instances`	Number of container instances registered into the cluster

Inventory data

EOL NOTICE

After March 2022, we're discontinuing support for several capabilities, including inventory data for cloud integrations. For more details, including how you can easily prepare for this transition, see our Explorers Hub post.

Inventory data provides information about the service's state and configuration. For more about inventory data, see Understand and use data.

aws/ecs/cluster

Name	Description
`status`	The status of the cluster. The valid values are `ACTIVE` or `INACTIVE`. `ACTIVE` indicates that you can register container instances with the cluster and the associated instances can accept tasks.
`name`	User-generated string to identify the cluster.
`awsRegion`	AWS region where the cluster is running.

aws/ecs/service

Name	Description
`status`	The status of the service. The valid values are `ACTIVE`, `DRAINING` or `INACTIVE`. `ACTIVE` means the instance accepts new tasks, `DRAINING` means the instance prevents new tasks from being started and notifies the service scheduler to move tasks to other instances in the cluster, generally used with the purpose of maintaining the instance or scale it down and `INACTIVE` means the instance is not active.
`clusterName`	User-generated string to identify the cluster.
`serviceName`	User-generated string to identify the service.
`launchType`	Type of infrastructure on which tasks and services are hosted. The valid values are `EC2` and `FARGATE`.
`awsRegion`	AWS region where the service is running.
`deploymentMaximumPercent`	Upper limit on the number of service's tasks that are allowed in the `RUNNING` or `PENDING` state during a deployment, as a percentage of the `desiredCount`.
`deploymentMinimumPercent`	Lower limit on the number of service's tasks that must remain in the `RUNNING` state during a deployment, as a percentage of the `desiredCount`.
`desiredCount`	The number of instantiations of the specified task definition to place and keep running on the cluster.
`taskDefinition`	ARN of the task definition file that describes the containers that form the application.