Full-Stack Observability: One Tool to Rule Them All
summary
Demonstrates a working full‑stack pipeline: frontend (JS) and backend (Java) use automatic instrumentation via a Java agent JVM parameter and OpenTelemetry JS libraries, all sent to an OpenTelemetry Collector that receives, batches, processes and exports traces/metrics/logs to Jaeger, OpenObserve and Signals frontends. Built demo microservices (frontend, product, order, user) produce waterfall traces that reveal a 500 error and exception logs; session replay works via session replay export. Notes: 1–2% Java overhead, sampling to control storage costs, collector queue drops when full.
AI-written summary · about
chapters (15)
- 0:00Introduction and motivation
- 3:59Problems with current tools
- 7:29What is observability?
- 9:19Metrics, logs and traces
- 13:14Full stack observability solution
- 15:40OpenTelemetry instrumentation
- 19:10OpenTelemetry collector
- 24:48Data storage and setup
- 25:09Key features of observability
- 30:33Demo setup overview
- 31:59Demo: Instrumentation in action
- 36:14Demo: Observability tools
- 41:37Challenges of implementation
- 44:11Recommendations and wrap-up
- 45:40Q&A session
related talks
01h 10mObservability For (Java) Devs - 2023 Edition
58mThrough the looking glass: Effective observability for cloud native applications
49mDistributed Tracing in Java? OpenTelemetry for the Rescue!
35mOpenTelemetry for Busy People
47mA Practical Introduction to Open Telemetry
47mThrough the Looking Glass: Effective Observability for Cloud-Native Applications
37mMake your Java application observable with no code changes using OpenTelemetry
51mPractical introduction to OpenTelemetry tracing