Logging, Instrumentation, Dashboards and Alerts - for Developers
summary
Live demo of a URL-shortener built on Azure with Functions, a queue, and a SQL DB shows instrumentation end-to-end: SDK + TelemetryClient emits custom metrics to Application Insights, logs flow through NLog and Seq locally, and alerts use Azure Monitor and ARM templates. Key findings: sampling drops some logs, custom metrics take ~5 minutes to appear, rollups compress older data, and watching percentiles not averages reveals spikes (max queue delay > seconds) that averages hide.
AI-written summary · about
chapters (11)
- 0:00Introduction and speaker intro
- 1:56Theory behind monitoring and logging
- 8:02Architecture for metrics and logs
- 12:19Why developers should care
- 14:04What to monitor in applications
- 23:46Demo: Infrastructure metrics in Azure
- 30:41Demo: Application metrics in Azure
- 34:56Demo: Querying logs in Azure
- 37:47Demo: Creating alerts from logs
- 40:20Limitations of Application Insights
- 42:47Q&A and resources
related talks
38mLogging, Instrumentation, Dashboards, Alerts
01h 01mLogging, tracing and metrics: instrumentation in .NET and Azure
52mWhen to use what monitoring service
52mMonitoring and alerting like a pro with Azure Monitor/Application Insights
49mMonitoring real-life Azure applications: When to use what and why
43mMoving Up the Monitoring Stack
01h 00mDon't Fly Blind - Implementing Effective Application Instrumentation
01h 00mMonitoring a Global Multitenant Service