Reduce Risks With Continuous...
summary
Spinnaker pipelines gain automated verification that compares a canary to a baseline using metric analysis and a numeric score so rollouts fail fast rather than relying on slow manual checks. The team demonstrates Kayenta canary scoring against Prometheus and APM, an NLP log diff that surfaces repeated errors, and an Optimax prototype that produces pass fail results at Netflix scale. Best practices include using 5–6 strong application metrics, ensuring equal traffic to canary and baseline, and adding in-stage health checks like rolling red black with in-stage decisions and alert/alerting hooks for downstream impact.
AI-written summary · about
chapters (9)
- 0:00Start
- 8:07What Is Goodness Verification
- 11:31Best Practices for Canary
- 13:46Try To Use Application Metrics When Possible
- 16:12Web Hooks
- 16:57Manual Checks
- 23:12Future
- 26:12Alerting Systems
- 28:15Operational Validation
related talks
51mCutting edge Continuous Delivery: Automated Canary Analysis through Spring based Spinnaker
42mMetrics Driven Blue-green Deployments using Spinnaker’s Cloud...
35mCutting-edge Continuous Delivery: Automated Canary Analysis on CF using Spinnaker
20mAutomated Verification of Deployments for Risk Assessment
34mEnabling Invisible Infrastructure Upgrades with Automated Canary Analysis
12mImplementing Advanced Deployment Using Spinnaker and Canary
21mContinuous Verification Updates
14mCanary Deployments for Infrastructure Upgrades