Power your .NET application with the new generation of diagnostics
summary
Automates safe production diagnostics by using a sidecar agent that subscribes to EventPipe event counters and triggers quick process snapshots (around 100–200 ms, demo shows 185 ms) instead of heavy dumps. Uses DiagnosticClient to listen for memory, CPU and first‑chance exceptions, then inspects snapshots programmatically with clrmd to locate a large static byte[] leak, root path and assembly load‑context holders. Also demonstrates custom‑header triggers and .NET 5 reverse‑attach for early bootstrap tracing.
AI-written summary · about
chapters (15)
- 0:00Intro
- 2:15Goal
- 6:26Troubleshooting
- 8:30Challenges
- 10:21Scenario
- 16:39Create your own
- 18:09Demo
- 21:44Snapshot
- 28:54Query
- 33:49Modules
- 36:09AssemblyLoadContext
- 38:54Diagnostic Client
- 40:44Custom Event Counter
- 42:32Reverse Channel Communication
- 46:55Assembly Load Context
related talks
55mPreemptive monitoring applications in production
01h 04mAdvanced .NET debugging techniques from real world investigations
53mDebugging .NET Apps
43mSelf-Aware Applications: Automatic Production Monitoring
01h 02mFrom ETW to EventPipe
01h 05mDiagnosticSourcery 101
03mInterview Speaker
01h 01mLogging, tracing and metrics: instrumentation in .NET and Azure