Health checks

Confirm a workspace and its clusters are healthy - what to check in the Console and, for In-VPC, in your Kubernetes cluster.

Use these checks to confirm a workspace and its clusters are healthy, whether after a change or as a routine review.

In the Console

Cluster status. The Clusters list shows each cluster's state - Running, Suspended, Pending, Error, or Degraded. A healthy cluster is Running (or intentionally Suspended). Hover over an Error state for the reason; see Lifecycle states and error codes.
Activity history. Each cluster's recent lifecycle events (create, resume, suspend, update, scale) with outcomes - the first stop when something looks off.
Run History. A spike in failed or timed-out queries is an early signal. Filter by status to spot patterns; see Query history.
Version upgrades. Confirm platform components and the Metadata Service are on expected versions - see Upgrades and releases.

For In-VPC deployments, also confirm the workspace pods and platform components are running in your Kubernetes cluster:

kubectl get pods -n <WORKSPACE_NAME>
kubectl get pods -n e6operator

Expected workspace pods include console, envoy, xds, and the metadata services. See the In-VPC run-first-query checks for AWS and Azure.

Clusters are Running or intentionally Suspended; none stuck in Pending, Error, or Degraded.
Failed-query and timeout rates are within your normal baseline.
Suspended clusters resume on the first query.