From Kernel To Cloud: Building a Complete Open Source Observ...
summary
Claims a full open source pipeline that captures kernel-level telemetry and surfaces root causes; uses Inspektor Gadget eBPF gadgets (CPU profile, OOM kill, TCP) running in Docker Compose, exports OTLP to OpenTelemetry Collector/Data Prepper, indexes into OpenSearch and visualizes with PPL dashboards. Live demo shows CPU hot threads from kernel stacks, out‑of‑memory kills with PIDs, and TCP source/destination metrics; the gadgets produce logs and metrics that successfully reveal issues invisible to application logs.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 1:03OpenTelemetry for observability
- 3:32OpenSearch dashboards for observability
- 8:12Kernel-level data in observability
- 9:15Introduction to eBPF and debugging
- 11:51eBPF tools for observability
- 15:23eBPF with OpenTelemetry demo
- 16:24Debugging high CPU usage
- 20:41Detecting cache misses
- 22:48Monitoring website traffic
- 24:14Live dashboards with eBPF
- 25:58Key takeaways and resources for further learning
- 27:01OpenSearch observability community
- 28:05Q&A
related talks
23mOpenTelemetry Auto Instrumentation for Effortless Observability
55mCloud-Native Observability with OpenTelemetry
45mSimplifying Instrumentation in complex systems at scale with Open Telemetry
11mObservability 2.0 - Much more than just logs,metrics and traces
50mGuide to Observability with OpenTelemetry
02h 02mHow To Troubleshoot and Monitor Applications using OpenTelemetry
23mModern Application Debugging: An Intro to OpenTelemetry
32mDistributed Tracing in Action: Our Journey with OpenTelemetry