Question 840
During a post-incident review at a logistics office, the original problem was: A branch SLA oscillates between healthy and failed every few seconds because recovery thresholds are too sensitive near its jitter boundary. Which preventive practice most directly addresses its cause?
Explanation
Correct Answer
Tune monitoring hysteresis with real noisy-link data before production deployment. This control reduces recurrence of the documented mechanism: Over-sensitive probe state transitions cause steering instability around normal variation. It creates a repeatable check around the needed response: Adjust supported threshold, fail/recover criteria or hysteresis behavior using measured path stability.
Incorrect Answers
Baseline latency distributions and application tolerances before setting targets. That safeguard addresses another potential failure: A performance SLA is configured with a 5-ms latency threshold despite a normal 40-ms regional WAN path, excluding all members under healthy conditions. It is not the most targeted recurring control for the described cause: Over-sensitive probe state transitions cause steering instability around normal variation.
Verify probe source IP and provider return routes during link commissioning. That safeguard addresses another potential failure: The performance SLA reports high loss on a backup line because probes use a source address not routed by its provider. It is not the most targeted recurring control for the described cause: Over-sensitive probe state transitions cause steering instability around normal variation.
Include DNS health in monitoring target configuration and outage triage. That safeguard addresses another potential failure: A remote edge reports a member unreachable after DNS changes because its health-check server is configured as a hostname and the FortiGate cannot resolve it. It is not the most targeted recurring control for the described cause: Over-sensitive probe state transitions cause steering instability around normal variation.