Question 210
A support engineer assumes every anomaly is a device-down alarm even when the affected rule is a tenant connectivity expectation. Which initial check would most directly distinguish the likely cause?
Explanation
Correct Answer
This checks the relevant evidence first: inspect anomaly type and affected graph nodes before selecting the next technical investigation. The underlying concept is that Apstra anomaly categories distinguish service intent violations, device issues and analytics-detected deviations. After confirming the finding, a controlled remediation is to route the issue to the correct service, telemetry or device workflow based on the observed anomaly evidence.
Incorrect Answers
That choice—“Inspect the specific service anomaly, affected objects and intended virtual-network configuration and state”—addresses this other problem: A blueprint raises a service anomaly for a virtual network even though all interfaces in its rack are administratively up. Here the scenario is: A support engineer assumes every anomaly is a device-down alarm even when the affected rule is a tenant connectivity expectation. The relevant inspection is: Inspect anomaly type and affected graph nodes before selecting the next technical investigation. The most discriminating first check is “Inspect anomaly type and affected graph nodes before selecting the next technical investigation”.
The option “Inspect the probe's monitored metric, stage, threshold and underlying samples before correlating service state” is useful in a different setting: A custom telemetry probe flags a latency anomaly but the blueprint shows no service configuration errors. The case being tested is: A support engineer assumes every anomaly is a device-down alarm even when the affected rule is a tenant connectivity expectation. The relevant inspection is: Inspect anomaly type and affected graph nodes before selecting the next technical investigation. The most discriminating first check is “Inspect anomaly type and affected graph nodes before selecting the next technical investigation”.
“Inspect object types, graph relationship properties and returned identities for the intended rack” is relevant to a different condition: An operator wants to identify managed nodes by rack and role in the blueprint graph. The stated question instead concerns: A support engineer assumes every anomaly is a device-down alarm even when the affected rule is a tenant connectivity expectation. The relevant inspection is: Inspect anomaly type and affected graph nodes before selecting the next technical investigation. The most discriminating first check is “Inspect anomaly type and affected graph nodes before selecting the next technical investigation”.