L17. Configuring DevOps Monitoring: Azure Monitor, Application Insights & GitHub Insights
Application Insights, VM Insights, and Container Insights are each tuned to what actually matters for that resource type, and GitHub Insights covers repository-level activity Azure Monitor never sees.
Azure Monitor: The Umbrella Platform
Azure Monitor is Microsoft's unified observability platform for Azure resources. It collects two primary types of telemetry: metrics (numerical measurements over time, like CPU utilization) and logs (detailed events and traces stored in a Log Analytics workspace). Queries against these logs use Kusto Query Language (KQL), a powerful tool we cover in the next lesson.Azure Monitor's value lies in its integration with DevOps tooling. You can automatically create work items in Azure DevOps from alerts, gate a pipeline stage on an alert's active state, or send notifications to Teams when monitoring thresholds breach. This closes the loop between observability and incident response.
Application Insights: APM for Your Code
Application Insights is Azure Monitor's application performance monitoring (APM) component. It is instrumented directly into your application (via SDKs for .NET, Java, Python, Node.js, etc.) and tracks:- Request rates and response times (how fast is your app?)
- Failure rates and exception stack traces (what is breaking?)
- Dependency call performance (is the database slow?)
- User behavior and custom events (what are users actually doing?)
Unlike infrastructure monitoring (which watches CPU and memory), Application Insights watches your code's behavior from the user's perspective. The exam tests recognizing that Application Insights is the purpose-built tool for application-level telemetry, not VM Insights or generic log queries.
Pre-Built Insights: Purpose-Built for Each Resource Type
Azure Monitor offers several pre-built "insights" experiences tailored to specific resource types:
| Insight Type | What It Monitors | Why It Matters |
|---|---|---|
| VM Insights | CPU, memory, disk, network, process dependencies | Diagnose whether infrastructure limits are causing performance issues |
| Container Insights | Kubernetes cluster health, pod resources, container logs | Monitor containerized applications without hand-building every dashboard |
| Storage insights | Latency, throughput, request patterns | Understand storage performance and capacity trends |
| Network insights | Connection health, packet loss, latency | Diagnose network connectivity and throughput issues |
GitHub Insights: Repository and Workflow Analytics
GitHub Insights provides built-in repository analytics independent of Azure Monitor. It tracks pulse (activity overview), contributors (who is committing?), traffic (page views and clones), commits (commit frequency), and code frequency (lines changed per week). GitHub Actions workflows also have an insights view showing run history, duration, and success rate.Scoping Telemetry Collection
The exam emphasizes a key principle: deliberately scope telemetry collection per resource type rather than maximizing collection everywhere. Indiscriminately turning on every possible metric and log increases Log Analytics ingestion cost and alert noise without necessarily improving detection quality. A mature monitoring strategy selects what to collect based on each resource's operational needs.
- ✓Application Insights is Azure Monitor's application performance monitoring component, instrumented into an application to collect request rates, response times, failure rates, dependency call performance, and exceptions
- ✓VM Insights, Container Insights, Storage insights, and Network insights are pre-built Azure Monitor experiences, each tuned to the telemetry that is actually meaningful for that specific resource type
- ✓GitHub Insights provides built-in repository analytics (pulse, contributors, traffic, commit/code frequency), separate from Azure Monitor, for GitHub-hosted repository and workflow activity
- ✓Azure Monitor can integrate with DevOps tooling, such as automatically creating a work item from an alert or gating a pipeline stage on an alert's active state
- ✓Deliberately scoping telemetry collection per resource type (rather than maximizing collection everywhere by default) controls both Log Analytics ingestion cost and alert noise, without necessarily improving detection quality
1. A team wants to measure their web application's request rates, response times, and dependency call performance, and capture unhandled exceptions with stack traces. Which Azure Monitor component is purpose-built for this?
2. Why does Azure Monitor offer separate pre-built experiences like VM Insights and Container Insights rather than one generic monitoring view for every resource type?
3. A team wants visibility into their GitHub repository's commit frequency, contributor activity, and traffic trends, separate from any Azure-hosted telemetry. Where should they look?
Recommended: Pluralsight
This free course covers the theory. Pluralsight adds structured DevOps Engineer Expert learning paths, hands-on Azure Pipelines and GitHub Actions labs, and timed practice exams to make it stick before exam day.