AZ-104 Deploy and Manage Azure Compute Practice Question
A company runs a stateless web tier on a scale set of Azure virtual machines. The operations team wants the scale set to add instances automatically when average CPU utilization exceeds 70 percent for 10 minutes, and to remove instances when it drops below 30 percent. They also need to ensure that instances are not removed during a 15-minute deployment window. Which Azure feature should you configure?
⚠ Common exam trap
Many candidates confuse load balancer health probes or alert-driven functions with native autoscale, which is the only feature that directly provides metric-based scale set scaling and instance protection.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Virtual machine scale set autoscale rules combined with scale set instance protection.
Virtual machine scale set autoscale rules evaluate metrics like CPU over a defined time window and adjust instance count up or down. Instance protection policies can be applied to specific instances so the scale set does not remove them during operations such as deployments. Configuring both features together provides automatic metric-based scaling and the temporary protection the operations team requires.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Virtual machine scale set autoscale rules combined with scale set instance protection.
Why this is correct
Autoscale rules in a virtual machine scale set monitor metrics such as CPU and adjust instance count when thresholds are breached. Instance protection policies, specifically Protect from scale-in, prevent the scale set from removing protected instances during operations like deployments. Together they deliver the metric-based scaling and the temporary removal protection the team needs.
- ✗
Azure Monitor alerts that trigger an Azure Function to resize the scale set.
Why it's wrong here
Azure Monitor alerts can notify or trigger actions, but using a function to resize a scale set is a custom orchestration approach that is harder to maintain and does not include built-in instance protection. The native autoscale engine already handles metric evaluation and scaling, making this option unnecessarily complex and incomplete for the deployment-window requirement.
- ✗
Azure Automation runbooks that start and stop VMs based on a schedule.
Why it's wrong here
Runbooks are schedule-based and can start or stop individual VMs, but they do not natively evaluate CPU metrics against thresholds, nor do they integrate with scale set instance protection during deployments. Using them would require custom scripting and would not provide the automatic metric-driven scaling and protection behavior the scenario requires.
- ✗
Azure Load Balancer health probes that restart unhealthy instances automatically.
Why it's wrong here
Health probes determine whether an instance should receive traffic, and they can trigger replacement of unhealthy instances only when paired with specific scale set settings. They do not evaluate CPU thresholds to add or remove capacity, and they cannot enforce a 15-minute no-removal window during deployments, so they do not meet the scaling requirement.
Go deeper
Related to this question
About these practice questions
This AZ-104 question is part of Courseiva's 1,053-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Microsoft exam blueprint
This AZ-104 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AZ-104 exam.