Reduce Risks with Continuous...
summary
Spinnaker already runs automated canaries using Kayenta and Autopilot to compare two datasets of metrics and return pass/fail scores, and teams extend that with Prometheus or New Relic queries and a referee UI. Best practices call for 5–6 metrics (errors carry more signal than successes) and avoid the 600–900‑metric trap. Optimax adds log NLP to surface repeated critical messages and API‑level comparisons, and the next step is embedding rolling red/black health hooks so deployments can pause, rollback, or consult alerting systems mid‑rollout.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 1:36About App
- 2:20Verification at speed
- 3:40Manual verification at speed
- 5:11Verification at every stage
- 7:52Good news modification
- 11:19Best practices for Canary
- 14:00Comparing two data sets
- 14:38Metrics
- 15:54Canary and Beyond
- 17:39Canary Technology
- 22:58The Future
- 23:13Checkpoints
- 24:23Thresholds
- 27:35Ecosystem
- 28:46Related Sessions
- 29:53Wrapup
related talks
20mAutomated Verification of Deployments for Risk Assessment
42mMetrics Driven Blue-green Deployments using Spinnaker’s Cloud...
51mCutting edge Continuous Delivery: Automated Canary Analysis through Spring based Spinnaker
14mCanary Deployments for Infrastructure Upgrades
22mAnalysis in Action: Real-World Metrics Driven Argo Roll...
52mHolistic Monitoring
37mAutomatically Detect the Top Performance & Scalability Issues in Distributed
20mUsing Monitoring to Speed Up Your DevOps Journey