Quick start: OTel Container Insights on Amazon EKS
This guide walks you through enabling OTel Container Insights on an existing Amazon EKS cluster. By the end of this procedure, your cluster sends infrastructure metrics and container logs to Amazon CloudWatch with Enhanced Observability enabled.
You can enable OTel Container Insights in two ways: by using the AWS Management Console (fastest) or by
using the AWS CLI. Both approaches install the same
amazon-cloudwatch-observability EKS add-on with the OTel Container Insights
configuration. You don't need manual agent deployment, Helm charts, or custom collector
pipelines. The entire process takes under 5 minutes.
Prerequisites
Before you enable OTel Container Insights, verify that you meet the following requirements.
-
An existing Amazon EKS cluster running Kubernetes version 1.28 or later
-
Platform version
eks.1or later -
Version 6.2.0 or later of the
amazon-cloudwatch-observabilityadd-on -
AWS CLI version 2.15.0 or later (for CLI-based setup)
-
kubectlconfigured to communicate with your target cluster -
IAM permissions:
eks:CreateAddon,eks:DescribeAddon, andiam:CreateServiceLinkedRole -
The EKS Pod Identity Agent add-on installed on your cluster, or IAM Roles for Service Accounts (IRSA) configured
-
Outbound internet access from the cluster to CloudWatch endpoints
Enable OTel Container Insights (console)
The AWS Management Console provides the fastest path to enable OTel Container Insights.
To enable OTel Container Insights by using the console
-
Open the Amazon EKS console at https://console.aws.amazon.com/eks/
. -
Choose Clusters, and then choose your cluster name.
-
Choose the Observability tab.
-
Choose Enable Container Insights and follow the on-screen instructions.
For a detailed console walkthrough, see Enable OTel Container Insights from the console.
Enable OTel Container Insights (AWS CLI)
Use the following steps to enable OTel Container Insights by using the AWS CLI.
Step 1: Create the IAM role
Create an IAM role that allows the CloudWatch Observability add-on to send data to CloudWatch.
To create the CloudWatch Observability add-on IAM role
-
Run the following command to create the role with a trust policy for EKS Pod Identity.
aws iam create-role \ --role-name EKS-CloudWatch-Observability-Role \ --assume-role-policy-document '{ "Version": "2012-10-17", "Statement": [{ "Effect": "Allow", "Principal": { "Service": "pods.eks.amazonaws.com" }, "Action": ["sts:AssumeRole", "sts:TagSession"] }] }' -
Attach the
CloudWatchAgentServerPolicymanaged policy to the role.aws iam attach-role-policy \ --role-name EKS-CloudWatch-Observability-Role \ --policy-arn arn:aws:iam::aws:policy/CloudWatchAgentServerPolicy
Step 2: Create the Pod Identity association
Associate the IAM role with the CloudWatch agent service account in your cluster.
To create the Pod Identity association
-
Run the following command. Replace
cluster-namewith the name of your Amazon EKS cluster andaccount-idwith your AWS account ID.aws eks create-pod-identity-association \ --cluster-namecluster-name\ --namespace amazon-cloudwatch \ --service-account cloudwatch-agent \ --role-arn arn:aws:iam::account-id:role/EKS-CloudWatch-Observability-Role
Step 3: Install the Amazon CloudWatch Observability add-on
Install the amazon-cloudwatch-observability add-on with OTel Container
Insights enabled.
To install the add-on
-
Run the following command. Replace
cluster-namewith the name of your Amazon EKS cluster.aws eks create-addon \ --cluster-namecluster-name\ --addon-name amazon-cloudwatch-observability \ --configuration-values '{"otelContainerInsights":{"enabled":true}}'Important
The
otelContainerInsights.enabledconfiguration is required. OTel Container Insights is not enabled by default.
Step 4: Verify the add-on status
Confirm that the add-on installed successfully.
To verify the add-on status
-
Run the following command. Replace
cluster-namewith the name of your Amazon EKS cluster.aws eks describe-addon \ --cluster-namecluster-name\ --addon-name amazon-cloudwatch-observability \ --query "addon.status" \ --output textThe expected output is
ACTIVE.
Step 5: Confirm agent pods are running
Verify that the CloudWatch agent pods are running in the
amazon-cloudwatch namespace.
To confirm agent pods are running
-
Run the following command.
kubectl get pods -n amazon-cloudwatch -l app.kubernetes.io/name=cloudwatch-agentAll agent pods must show
Runningstatus.
Verify data in CloudWatch
After you complete the setup, Container Insights data appears in CloudWatch within 3 to 5 minutes.
Check metrics
To check metrics in CloudWatch
-
Open the CloudWatch console at https://console.aws.amazon.com/cloudwatch/
. -
In the navigation pane, choose Query Studio.
-
Search for metrics such as
container_cpu_usage_seconds_totalby using PromQL.
Check logs
To verify that log groups exist for your cluster, run the following command. Replace
cluster-name with the name of your Amazon EKS cluster.
aws logs describe-log-groups \ --log-group-name-prefix "/aws/containerinsights/cluster-name" \ --query "logGroups[].logGroupName" \ --output table
Expected time-to-data
The following table shows the expected latency for each signal type after you enable OTel Container Insights.
| Signal | Expected latency |
|---|---|
| Infrastructure metrics | 2–3 minutes |
| Container logs | 2–3 minutes |
| Performance log events | 3–5 minutes |
Troubleshooting
Use the following guidance to resolve common issues when you enable OTel Container Insights on Amazon EKS.
Add-on status shows CREATE_FAILED or DEGRADED
Symptom: When you run aws eks
describe-addon, the status shows CREATE_FAILED or
DEGRADED.
Cause: The add-on installation failed, typically because of insufficient IAM permissions or a missing Pod Identity association.
Solution: Complete the following steps to resolve this issue.
-
Run the following command to check for detailed error information. Replace
cluster-namewith the name of your cluster.aws eks describe-addon \ --cluster-namecluster-name\ --addon-name amazon-cloudwatch-observability \ --query "addon.health" -
Verify that the IAM role exists and has the
CloudWatchAgentServerPolicyattached. -
Verify that the Pod Identity association targets the correct namespace (
amazon-cloudwatch) and service account (cloudwatch-agent). -
Delete the failed add-on and reinstall it after you resolve the issue.
aws eks delete-addon \ --cluster-namecluster-name\ --addon-name amazon-cloudwatch-observability
Agent pods are in CrashLoopBackOff or Pending state
Symptom: When you run kubectl get pods -n
amazon-cloudwatch, one or more pods show CrashLoopBackOff or
Pending status.
Cause: The agent pods can't start because of insufficient node resources, missing permissions, or network connectivity issues.
Solution: Complete the following steps to resolve this issue.
-
Check the pod events for detailed error messages.
kubectl describe pod -n amazon-cloudwatch -l app.kubernetes.io/name=cloudwatch-agent -
Check the agent container logs for startup errors.
kubectl logs -n amazon-cloudwatch -l app.kubernetes.io/name=cloudwatch-agent --tail=50 -
Verify that your nodes have sufficient CPU and memory available for the agent pods.
-
Verify that the EKS Pod Identity Agent add-on is installed and running.
kubectl get pods -n kube-system -l app.kubernetes.io/name=eks-pod-identity-agent
Metrics not appearing in CloudWatch after 5 minutes
Symptom: The agent pods show
Running status, but no metrics appear in CloudWatch after 5 minutes.
Cause: The agent can't send data to CloudWatch, typically because of network restrictions or incorrect IAM permissions.
Solution: Complete the following steps to resolve this issue.
-
Verify that the agent pods can reach CloudWatch endpoints. Check that your VPC security groups and network ACLs allow outbound HTTPS traffic (port 443) to CloudWatch endpoints.
-
Check the agent logs for permission errors or connection timeouts.
kubectl logs -n amazon-cloudwatch -l app.kubernetes.io/name=cloudwatch-agent --tail=100 | grep -i "error\|timeout\|denied" -
Verify that the IAM role has the
CloudWatchAgentServerPolicypolicy attached and that the trust policy allowspods.eks.amazonaws.com. -
If you use a VPC endpoint for CloudWatch, confirm that the endpoint policy allows the required actions.