Shifting Left on Production Excellence with Observability
summary
Argues that shifting production left happens by capturing rich event data, using distributed tracing and event-based SLOs to turn telemetry into business decisions. Describes tools like otelcli and OpenTelemetry, shows SLO math (99.9% gives a 0.1% error budget — ~1,000 bad requests per million per month) and alerts on burn rate, demonstrates tracing that links an upstream change to downstream user impact, and reports fast mitigation when that signal exists.
AI-written summary · about
chapters (15)
- 0:00Start
- 1:37Shelby Spees
- 2:55Developers Should Be on Call
- 4:02Novel Emergent Failure Modes
- 6:36Observability
- 7:31Observability Is for the Hard To Debug Problems
- 9:46Making the Business Case for Production Excellence
- 10:04Service Level Objectives
- 12:54Error Budget
- 16:19How To Bring Production Excellence to Your Organization
- 19:12Next Generation Developers
- 20:27The Common Roadblocks
- 22:55Quick Wins
- 23:11Instrumenting Your Builds
- 32:37Line between Observability and Regular Monitoring
related talks
47mCultivating Production Excellence
58mCultivating Production Excellence
29mCultivating Production Excellence
25mHow to detangle and understand production
34mObservability 2.0: Shifting Left From Reactive Mon...
43mGo in DevOps Panel
29mObservability; a gentle introduction
32mThe Role of Observability in Cloud Native Environments