Murphy’s Laws for observability
summary
Claims that observability fails when you under-sample and lose rare but critical events, showing sampled traces can appear 1–2s while full capture reveals 29–40s slowdowns. Recommends unified collection with OpenTelemetry to correlate metrics, traces, and logs across scale—examples include systems with 700 active services or Lambda 1.2s average durations—and warns that low-resolution ingestion, time skew, and retention choices break alerts and hide outages.
AI-written summary · about
chapters (7)
- 0:00Intro
- 0:57Murphys Laws
- 4:47Every Problem Creates New Problems
- 9:00Nothing is as Easy as it Looks
- 12:03Things Get Worse Under Pressure
- 24:12Open Telemetry
- 26:26Ashley Perry Statistical Axiom
related talks
38mThe Murphy's Laws of Observability
32mRules of Observability: The Data Edition
26mYou Can’t Fix What You Can’t See: Tales from the Observability Frontier
11mObservability 2.0 - Much more than just logs,metrics and traces
29mObservability; a gentle introduction
39mThe Ticking Timebomb of Observability Expectations
39mObservability for Emerging Infra: What Got You Here Won't Get You There
53m.NET Rocks Live