After readiness failures, a flaky health endpoint flickers. Which probe field requires multiple consecutive successes before marking the Pod ready again?

For readiness probes, successThreshold configures how many consecutive successes are required after a failure before the container is considered ready again (liveness successThreshold is fixed at 1). Raising readiness successThreshold stabilizes flapping endpoints. failureThreshold counts consecutive failures, not recovery successes; periodSeconds and timeoutSeconds control cadence and per-probe timeouts, not the success streak.