3.4 - Service health, SLAs and service lifecycle
AZ-900 objective 3.4 covers monitoring, service health and the service lifecycle. Azure Monitor is the central platform that collects, analyzes and acts on telemetry - both metrics and logs - from Azure and on-premises resources, and Application Insights within it does application performance monitoring of a live app's response times, failure rates and dependencies. Azure Service Health gives a personalized, authenticated view of issues, planned maintenance and advisories that affect the subscriptions and regions you actually use, so during an incident it tells you whether a platform problem is hitting you. Azure Resource Graph queries a single, up-to-date inventory of resources such as every storage account across the tenant. Azure Advisor's five categories are cost, security, reliability, operational excellence and performance. On lifecycle, a service at general availability is released for production and backed by an SLA, while a public preview is for evaluation, not recommended for production, and typically has no SLA; a 99.99 percent SLA allows less downtime than 99.9. Expect scenarios that ask which monitoring service or lifecycle stage applies.
Central telemetry platform = Azure Monitor; live app performance = Application Insights. Personalized, authenticated view of platform incidents affecting you = Service Health. Query one inventory across the tenant = Resource Graph. GA has an SLA and is for production; public preview does not.
Practice questions
1. An administrator wants personalized recommendations to improve the cost, security, reliability, and performance of resources already deployed in a subscription. Which tool provides this?
- Azure Advisor (correct answer)
- Azure Service Health, which reports incidents in Azure regions
- Azure Monitor metrics charts configured for each resource
- The Pricing calculator filled in with the current usage volumes
Azure Advisor analyzes the resources in your subscription and gives personalized recommendations across cost, security, reliability, performance, and operational excellence. Service Health reports Azure platform incidents, Monitor charts raw telemetry without advice, and the Pricing calculator estimates costs before deployment.
2. Which Azure service is the central platform that collects, analyzes, and acts on telemetry (metrics and logs) from Azure and on-premises resources?
- Azure Monitor (correct answer)
- Azure Advisor, which only produces best-practice recommendation lists for resources
- Azure Service Health, which reports platform outages affecting the subscription
- Microsoft Purview, which discovers and classifies data across the whole estate
Azure Monitor is the central monitoring platform that collects metrics and logs from Azure and on-premises resources, then analyzes and acts on them. Advisor only recommends, Service Health reports platform incidents, and Purview governs data.
3. Within Azure Monitor, which data type consists of numeric values sampled at regular intervals, such as CPU percentage over time?
- Metrics (correct answer)
- Logs, which are timestamped records queried later with a dedicated query language
- Alerts, which notify a team when a monitored condition is met or crossed once
- Workbooks, which combine several visualizations into interactive shared reports
Metrics are numeric values sampled at regular intervals, ideal for near-real-time charts like CPU over time. Logs are timestamped records queried with KQL, alerts fire on conditions, and workbooks are visualization reports.
4. An engineer wants to run flexible queries across collected log data using Kusto Query Language (KQL). Which Azure Monitor component stores those logs and is queried this way?
- A Log Analytics workspace (correct answer)
- A metrics namespace holding only numeric time-series counters, not any logs
- An Application Insights availability test scheduled to run from several regions
- An Azure Service Health alert rule that watches for broad platform incidents
A Log Analytics workspace is where Azure Monitor stores log data, queried with Kusto Query Language (KQL). A metrics namespace holds numeric counters, an availability test checks endpoints, and a Service Health alert watches platform incidents - none is the log store queried with KQL.
5. A development team wants to monitor a live web application's response times, failure rates, and dependencies to detect performance problems. Which Azure Monitor feature is built for application performance monitoring?
- Application Insights (correct answer)
- A Log Analytics workspace used only to archive raw platform logs for the long term
- Azure Service Health, which tracks the health of the Azure platform itself by region
- Azure Advisor, which periodically suggests cost and reliability improvements to apply
Application Insights is the application performance monitoring (APM) feature of Azure Monitor: it tracks response times, failure rates, dependencies, and usage of live apps. A Log Analytics workspace stores logs, Service Health tracks the platform, and Advisor gives recommendations.
6. A team instruments its web app with Application Insights. Which TWO capabilities does it gain? (Choose TWO.)
- Live request rates, response times, and failure rates (correct answer)
- Detection of performance and availability problems (correct answer)
- Automatic negotiation of the monthly Azure invoice with the billing support team
- Physical replacement of failed disks in the Azure datacenter hardware when needed
- Guaranteed elimination of every possible bug in the app before the code is deployed
Application Insights surfaces live request rates, response times, and failure rates, and detects performance and availability problems in running apps. It does not negotiate invoices, replace hardware, or guarantee bug-free code.