The Trouble with Tracers
summary
Explains concrete failures and solutions: a mock tracer and a working demo of monkey-patching require and a custom shallow-clone for transpiled exports, shows how async callbacks and connection pooling produce misattributed spans, and why naive setTimeout wrapping leaks memory. Dives into async_hooks pitfalls, W3C Trace Context headers, OpenTelemetry merger and Elastic APM tooling, outlines partial fixes (require-in-the-middle, sampling, worker-thread offloading) and documents which hacks work today and which problems remain.
AI-written summary · about
chapters (9)
- 0:00Introduction to tracing
- 1:14Fundamentals of spans
- 2:55The challenge of automation
- 5:33Monkey patching techniques
- 10:48Handling asynchronous operations
- 13:06Memory and callback pitfalls
- 16:50Propagating trace context
- 19:12Evolution of tracing standards
- 22:57Best practices and future
related talks
22mTracing is for Everyone, Not Just Backend Engineers. (How Tracing Could Help Front-e...
35mFun with async hooks and stack traces
46mHow to Properly Blame Things for Causing Latency
27mTraces of Errors: Getting Better JavaScript Stacktraces
43mObservability for React applications
25mFollow The (Full)stack Trace: Understand Your App with Distributed Tracing
42mObservable Applications Using OpenTelemetry
30m"Observability for the Front End"