Configure ATP for Kubernetes

Preview

We're still working on this feature, but we'd love for you to try it out!

This feature is currently provided as part of a preview pursuant to our pre-release policies.

Learn how to install and configure the Adaptive Telemetry Processor (ATP) with the NRDOT Collector in Kubernetes environments. ATP reduces telemetry data volume while maintaining critical visibility into your applications.

Important

ATP for Kubernetes uses an experimental build. This entire build is experimental for Kubernetes installation, not just the ATP feature being in preview.

Installation methods

You can install the OpenTelemetry Collector with ATP using either Helm (recommended) or Kubernetes manifests:

The Helm installation method is the recommended approach for deploying ATP-enabled collectors in Kubernetes.

Before you begin

Ensure you have:

Your New Relic license key
kubectl configured to access your cluster
Cluster admin permissions

Download and customize values.yaml

Download the Helm chart values.yaml file:
bash
```
$curl -o values.yaml https://raw.githubusercontent.com/newrelic/helm-charts/master/charts/nr-k8s-otel-collector/values.yaml
```
Edit the values.yaml file with your specific configuration:
```
# Required: Set your cluster name and license key
cluster: "YOUR_CLUSTER_NAME"
licenseKey: "YOUR_NEW_RELIC_LICENSE_KEY"

# Enable ATP
enable_atp: true

# IMPORTANT: Use experimental collector image for ATP
images:
  collector:
    repository: "newrelic/nrdot-collector"  # Changed from nrdot-collector-k8s
    tag: "<NRDOT_COLLECTOR_LATEST_VERSION>"  # Use latest version from releases page
```
Find the latest NRDOT Collector version at the releases page and update the tag value.
For additional configuration options, see configuration parameters.

Install the Helm chart

Add the New Relic Helm repository:

bash

$helm repo add newrelic https://helm-charts.newrelic.com
$helm repo update

Install the chart using your customized values.yaml file:

bash

$helm upgrade nr-k8s-otel-collector newrelic/nr-k8s-otel-collector \
>  -f values.yaml \
>  -n newrelic \
>  --create-namespace \
>  --install

Verify the installation

Check that pods are running:
bash
```
$kubectl get pods -n newrelic --watch
```
You should see pods with names like nr-k8s-otel-collector-<hash> in the newrelic namespace.

Verify New Relic is receiving process data by running these queries in the query builder:

FROM Metric SELECT *
WHERE k8s.cluster.name='YOUR_CLUSTER_NAME'
AND metricName LIKE 'process.%'
LIMIT 100

FROM Metric SELECT uniqueCount(metricName)
WHERE k8s.cluster.name='YOUR_CLUSTER_NAME'
AND metricName LIKE 'process.%'

Access the ATP data in New Relic

Once the ATP is configured, it starts collecting data from your Kubernetes cluster. You can access this data in New Relic OpenTelemetry UI. For more information on New Relic OpenTelemetry UI, refer to OpenTelemetry APM UI.

To view the ATP process metric data in New Relic:

Go to one.newrelic.com > All Entities.
OR
Go to one.newrelic.com > Catalogs > Infrastructure.
Search for entity where you installed the NRDOT collector with ATP.
Select the entity then click Process in the left pane.
On the Process page, you can view all the processes running, along with their IDs, CPU and memory utilization metrics. It also displays the parent-child relationship between processes if a process spawns other processes.

The manifest installation method provides a way to set up OpenTelemetry for Kubernetes without using Helm. This method is suitable for users who prefer a more hands-on approach or have specific requirements that necessitate manual configuration.

Before you begin

Ensure you have:

Your New Relic license key
kubectl configured to access your cluster
Cluster admin permissions

Download and prepare manifest files

Copy the contents of the nr-k8s-otel-collector's rendered examples directory to your local workspace:

bash

$git clone https://github.com/newrelic/helm-charts.git
$cd helm-charts/charts/nr-k8s-otel-collector/examples/k8s-with-atp/rendered

Update the collector image in both daemonset.yaml and deployment.yaml files:
Change from:
```
image: docker.io/newrelic/nrdot-collector-k8s
```
To:
```
image: docker.io/newrelic/nrdot-collector:<NRDOT_COLLECTOR_LATEST_VERSION>
```
Important
This is an experimental build for Kubernetes installation. Find the latest version at the releases page.

Configure license key and cluster name

Update the secret.yaml file with your Base64-encoded New Relic license key:
```
data:
  licenseKey: <YOUR_BASE64_ENCODED_LICENSE_KEY>
```
To encode your license key:
- Linux/macOS: echo -n "<YOUR_LICENSE_KEY>" | base64
- Windows (PowerShell): [System.Convert]::ToBase64String([System.Text.Encoding]::UTF8.GetBytes("<YOUR_LICENSE_KEY>"))
Update your cluster name in both daemonset-configmap.yaml and deployment-configmap.yaml files:
Locate instances of k8s.cluster.name and replace <YOUR_CLUSTER_NAME> with your desired cluster name:
```
- key: k8s.cluster.name
  action: upsert
  value: YOUR_CLUSTER_NAME  # Replace with your cluster name
```

Deploy the manifests

Create the newrelic namespace and apply the manifests:

bash

$kubectl create namespace newrelic
$kubectl apply -n newrelic -R -f .

Verify the pods are running:
bash
```
$kubectl get pods -n newrelic --watch
```

Verify New Relic is receiving process data by running these queries in the query builder:

FROM Metric SELECT *
WHERE k8s.cluster.name='<YOUR_CLUSTER_NAME>'
AND metricName LIKE 'process.%'
LIMIT 100

FROM Metric SELECT uniqueCount(metricName)
WHERE k8s.cluster.name='<YOUR_CLUSTER_NAME>'
AND metricName LIKE 'process.%'

Access the ATP data in New Relic

To view the ATP process metric data in New Relic:

Go to one.newrelic.com > All Entities.
OR
Go to one.newrelic.com > Catalogs > Infrastructure.
Search for entity where you installed the NRDOT collector with ATP.
Select the entity then click Process in the left pane.
On the Process page, you can view all the processes running, along with their IDs, CPU and memory utilization metrics. It also displays the parent-child relationship between processes if a process spawns other processes.

ATP configuration options

When ATP is enabled, the following configuration is automatically applied:

Parameter	Description	Default value
`enable_atp`	Enable/disable ATP functionality. Set to `true` to enable ATP.	`false`
`images.collector.repository`	Must be changed to `newrelic/nrdot-collector` to use ATP.	`newrelic/nrdot-collector-k8s`
`images.collector.tag`	Image tag for ATP-enabled collector.	`<TO_BE_CONFIRMED>`

Troubleshooting

Learn how to troubleshoot ATP issues for your environment.

Query your data

Learn how to query ATP data in New Relic using NRQL.

Advanced-features

Learn how to enable advanced features for your ATP deployment.

Preview

Important

Installation methods .css-21sua1{background:none;border:none;width:0;padding:0;}

Before you begin

Download and customize values.yaml

Install the Helm chart

Verify the installation

Access the ATP data in New Relic

Before you begin

Download and prepare manifest files

Important

Configure license key and cluster name

Deploy the manifests

Access the ATP data in New Relic

ATP configuration options

Related resources

Troubleshooting

Query your data

Advanced-features

Installation methods