Real‑time health_status event tailing versus periodic inspect polling for low‑notification‑noise health alerts

Goal: Choose a monitoring pattern that delivers health state alerts with minimal noise while respecting the unknown latency between healthcheck execution and status visibility in Podman, especially under rootless operation.

Constraint: The solution must avoid frequent, redundant checks and handle potential event stream disconnections without introducing additional alert spam.

Uncertainty: The exact delay for status propagation from a healthcheck’s outcome to its appearance in podman events versus podman inspect is not documented, and rootless containers may exhibit different scheduling behavior that influences this lag.

What measurable latency exists between a healthcheck’s success/failure and its visibility in podman events vs. podman inspect for rootless containers?

How does intermittent event stream loss affect the reliability of health_status notifications compared to periodic inspect polling?