Monitoring
Monitor cluster health, query performance, and resource utilization from the Compute Plane, and export telemetry to your own stack.
e6data surfaces real-time cluster status and per-query metrics in the Compute Plane, and can export telemetry to your own observability stack. This page covers what you can see in-product; for exporting to external tools, see Observability export.
Cluster status
The Clusters list shows each cluster's current state:
Active
Cluster is running and serving queries
Suspended
Cluster is suspended (no compute, no cost)
Pending
Cluster is being created or resumed; not yet ready
Error
A problem prevented the cluster from starting
Hover over an Error status to see the underlying message. For specific error codes, see Cluster errors.
Cluster details
Click a cluster to see its connection info (endpoints, port, credentials reference - see Create and manage clusters), active queries, resource usage (executor count, CPU, memory), and recent activity.
Query metrics
Open Run History to see metrics per query - execution time, bytes scanned, rows returned, cluster used, and the user and access method (UI, JDBC, REST, and so on). Filter by user, cluster, status, or time range. See Query history and saved queries.
Alerting
Common alerts to set up in your observability stack:
Cluster fails to resume
More than 5 minutes in Pending after a resume is triggered
Query timeout rate spike
Sudden increase in timed-out queries
Concurrent query backlog
Queries queued longer than expected
Failed query rate
Above a baseline you've established for the cluster
See also
Observability export - export metrics, logs, and traces via OpenTelemetry.
Last updated
Was this helpful?