Debugging at Scale in Production - Deep into your Containers
summary
Shows how kubectl debug (introduced in v1.23) uses ephemeral containers to inspect crashed or bare‑bones pods and how KoolKits supplies prebuilt images (JVM, Go, Node, Python) so common tools are already available. Demonstrates an agent‑based developer observability workflow (Lightroom) that injects logpoints, nonbreaking snapshots, conditional captures for specific user IDs, and method duration metrics, plus tag‑based rollout to target many pods. Gives concrete commands, image names, and real demos rather than high‑level platitudes.
AI-written summary · about
chapters (21)
- 0:00Debugging at scale intro
- 1:28Speaker intro
- 2:30Debugging in Kubernetes
- 3:34Bare bones container issue
- 4:57Cube City Debug intro
- 5:40Traditional debugging demo
- 7:05Cube City Debug with ephemeral
- 8:11QuarkX intro
- 9:36QuarkX with Cube City Debug
- 10:40QuarkX debugging benefits
- 12:05Massive production debugging
- 13:32Developer observability
- 14:13Lightrun observability
- 15:16Log injection demo
- 17:47Snapshots demo
- 19:12Conditional snapshots
- 20:55Conditional snapshots demo
- 21:16Performance metrics
- 22:20Method duration metrics
- 22:42Debugging with tags
- 23:04Summary
related talks
34mDebugging at Scale in Production - Deep into Your Containers with Kubectl Debug
35mThe Art of Debugging Inside K8s Environment
40mDebugging in Golang Apps Inside K8s Environment
01h 00mDebug like a pro on Kubernetes
50mTroubleshooting & Debugging Microservices in Kubernetes
51mTroubleshooting & Debugging Production Microservices in Kubernetes
26mHow to Debug the Pod Which is Hard to Debug
31mDebugging Go: from zero to Kubernetes - Michele Caci