Probe completeness (quiescence)

Kubesonde probes continuously — it does not run once and stop. So "is it done?" is really "have the results stabilized enough to be worth reading?" Kubesonde answers this with quiescence.

What "complete" means

Probing is considered complete when the recorded probe count has been stable for a fixed window (30 seconds by default). In other words: if no new probes have been recorded for the whole window, the results have quiesced and it's a good time to fetch them.

This is a hint, not a hard stop. The controller keeps probing after it reports complete — connectivity can change, and Kubesonde keeps observing.

Where completeness shows up

On the resource status — as a complete flag and a Complete condition, so you can block on it:

kubectl wait --for=condition=Complete kubesonde/kubesonde-sample --timeout=300s

kubectl get kubesondes

NAME                 NAMESPACE   PROBES   COMPLETE   AGE
kubesonde-sample     default     228      true       5m

On the HTTP API — query it directly:

curl -s localhost:2709/probes/status

{
  "complete": true,
  "count": 228,
  "window": 30000000000,
  "secondsSinceLastChange": 42.5
}

• window is the quiescence window in nanoseconds — 30000000000 = 30 seconds.
• secondsSinceLastChange is how long the count has been unchanged. When it exceeds the window, complete becomes true.

Why wait for it

Fetching results immediately after applying a Kubesonde resource may return empty or incomplete data, because probing has barely started. Always wait for the Complete condition (or a few minutes, depending on the number of pods) before you fetch results.

TIP

The time to reach completeness scales with the number of pods and the number of probes. A large namespace takes longer; you can speed it up by raising probe concurrency.