Skip to main content

Node & pod health: is it the app or the node?

avuru obs team
Maintainers

The Nodes screen is live: CPU, memory and network per node, with the busiest pods underneath — collected by the sensor's node agent (kubeletstats), stored in the same ClickHouse as everything else.

  • Per-node utilization. Latest CPU (cores) and memory with inline trend sparklines, network rx/tx rates over the selected window, and pod counts.
  • Busiest pods. Click a node to scope the pod list; each pod shows its namespace, owning workload (deployment/statefulset/daemonset), CPU and memory.
  • One storage, one retention. Metrics land in dedicated ClickHouse tables with their own retention knob (retention.metrics, default 7 days) — no separate metrics system to operate.