One of several containers in a pod is marked as unhealthy after failing its livenessProbe many times.
Is this the action taken by the orchestrator to fix the unhealthy container?
The unhealthy container is restarted.
One of several containers in a pod is marked as unhealthy after failing its livenessProbe many times.
Is this the action taken by the orchestrator to fix the unhealthy container?
The unhealthy container is restarted.
This is the default behavior for failed liveness probes. The container is restarted with a back-off delay. The delay increases exponentially with each failure (10s, 20s, 40s, etc.). The delay is capped at 5 minutes.