Forget Production: Observability a Non-Production Necessity
summary
Claims that observability belongs in non-production by building impermanent non-production pipelines that spin up prod‑duplicates, slam them with 10x load, collect metrics with Dynatrace and Harness (plus GitOps/CI‑CD and chaos techniques), then destroy the environment. Shows an example pipeline that creates environments, validates performance and security gates, and claims it yields thousands of deployments and nearly a year of production with no failures, while cutting monitoring license costs. Watch if want concrete pipeline steps, tool names, and measurable outcomes.
AI-written summary · about
chapters (7)
- 0:00Non-production observability intro
- 1:06Why prod observability fails
- 3:58Redefining SDLC confidence
- 6:47Stress testing tenfold
- 8:09Impermanent non-prod environments
- 9:54Confidence pipeline example
- 12:20Production as confidence goal
related talks
20m"Testing" in Prod Isn't the Only Way: Why Observability Needs Structure
58mObserving Applications in Production
26mObservability and Its Role in the Software Life Cycle
53mObservability Driven Development
18mYour Toolchain is Production: The Case for Observability
58mCultivating Production Excellence
29mObservability; a gentle introduction
28mJourney to observability: tracking every function execution in production