Putting Chaos Into Continuous Delivery to Increase App...
summary
Adds LitmusChaos chaos stages into GitOps CD orchestrated by Keptn/Captain, using Locust load and Prometheus SLI/SLO evaluations to gate promotions. Chaos experiments are declarative CRs stored in Git, executed as pod delete during load; a single‑replica deployment fails the SLO (probe success <90–95%), a three‑replica deployment meets it, so changing deployment spec (not code) restores resilience. Production risk drops; Ketoki runs these experiments hundreds of times.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 2:09Why is chaos engineering important
- 3:40Assumptions
- 4:33Churn
- 5:35Hypothesis
- 6:18Declarative
- 7:10Litmus Chaos
- 8:25Presentation Overview
- 9:44Project Introduction
- 12:01Evaluation
- 14:27Use Case
- 17:32Flow
- 19:21Demo
- 23:55Key takeaways
related talks
30mUsing GitOps for Kubernetes Reliability at Scale
26mUsing GitOps to Increase System Resiliency with LitmusChaos
34mChaos Engineering For Hybrid Targets With LitmusChaos
30mBringing Order to Chaos: Make Your Systems More Resilient with Chaos Engineering
39mPracticing Cloud-Native Resliency Is a Challenge: Decoding It with Chaos...
29mCase Study: Bringing Chaos Engineering to the Cloud Native Develo...
28mCloud Native Chaos Engineering
28mChaos Engineering alongside Litmus and Jenkins