Run first query

Validate an In-VPC e6data install by verifying the cluster and workspace components, then running your first query.

Validate the install in two passes: confirm the Kubernetes components are healthy, then run a query.

Verify cluster components

echo "=== Cluster Components ==="

# Karpenter
kubectl get pods -n kube-system -l app.kubernetes.io/name=karpenter

# ALB Controller
kubectl get pods -n kube-system -l app.kubernetes.io/name=aws-load-balancer-controller

# cert-manager
kubectl get pods -n cert-manager

# e6-operator
kubectl get pods -n e6operator

# NodePools and EC2NodeClasses
kubectl get nodepools.karpenter.sh
kubectl get ec2nodeclasses.karpenter.k8s.aws

Verify workspace components

echo "=== Workspace Components (${WORKSPACE_NAME}) ==="

kubectl get pods -n ${NAMESPACE}
kubectl get namespaceconfig -n ${NAMESPACE} -o wide
kubectl get queryrouter -n ${NAMESPACE} -o wide
kubectl get monitoringservices -n ${NAMESPACE} -o wide
kubectl get svc -n ${NAMESPACE}

# Pod Identity associations
aws eks list-pod-identity-associations \
  --cluster-name ${CLUSTER_NAME}-eks --namespace ${NAMESPACE} --region ${REGION} \
  --query 'associations[].{SA:serviceAccount,Role:roleArn}' --output table

A healthy deployment shows the console, envoy, xds-control-plane, and o11y-collector pods Running, and the *-qr-envoy-external service with an external LoadBalancer address.

Run a query

1. Open the workspace's SQL Editor at your configured hostname.
2. Select your cluster and a catalog.
3. Run a simple query against a table you can see:

SELECT * FROM your_catalog.your_schema.your_table LIMIT 10;

If results return, your AWS In-VPC deployment is working end to end.

If something fails

See Troubleshooting for pods stuck pending, image pull errors, operator and reconciliation issues, Pod Identity problems, LoadBalancer provisioning, and monitoring.

Next

• Component versions and operations - ongoing operations and the workloads reference.
• Query editor - write and run queries.