An admin needs an alert when a specific VM's CPU exceeds 90% for 5 minutes, emailing the on-call group. Which components are required?
A.Activity log alert + runbook
Activity log alerts fire on control-plane events (create/delete) — not performance thresholds.
B.Metric alert rule on the VM + action group with email
Metric alert + action group = the standard pattern. Correct.
C.Log Analytics query alone
A query returns data but can't notify without an alert rule wrapping it.
D.Service Health alert
Service Health alerts are for Azure platform issues, not VM metrics.
Why this answer
CPU% is a platform metric — a metric alert rule (signal + condition + window) fires on it, and an action group delivers notifications (email/SMS/webhook/ITSM). Alert rules and action groups are separate objects — the rule defines WHEN, the action group defines WHO/HOW.
Correct answer: B · Last reviewed 2026-02-14
Official AWS references
Practice this interactively
Answer without seeing the solution first in the Monitoring & Backup practice session.
Practice Monitoring & Backup