Incremental Rollouts and Rollbacks with Business-Metric Control
summary
Covers strategies for incremental deployments and rollbacks with business-metric gating, examining deployment types, common problems, and how to choose metrics to monitor during releases. Uses Argo Rollouts to demonstrate monitoring of technical and business metrics, analyzers, forecasting, and automating rollback rules for canary-style releases.
AI-written summary · about
chapters (17)
- 0:00Intro. Uspacy
- 2:44Evolution of Deployments
- 7:21"We want more control"
- 8:19Releases
- 9:18Problematics
- 12:15What do we want?
- 13:09Our path to Canaries: how it was and how it became
- 17:16What did we consider?
- 19:40Why ArgoRollouts?
- 20:37What is ArgoRollout?
- 21:18Available analytics
- 23:12Examples of analyzers
- 24:05What do we use? How does it look?
- 25:20Rollback rules
- 27:55Conclusions: pleasantries and complexities
- 30:09What's next?
- 31:48Q&A
related talks
22mAnalysis in Action: Real-World Metrics Driven Argo Roll...
26mProgressive Delivery Made Easy with Argo Rollouts
40mProgressive Delivery with Kubernetes and Argo Rollouts
23mStop Deploying Blind! Using Observability and Argo Rollouts to Light the Way
53mDeliver Results, Not Just Releases: Control & Observability in CD
10mLightning Talk: Open Source Canary Deployments with Application Metric Analysis
20mIntuitive Progressive Delivery Across Microservices in a Depe...
35mProgressive Delivery with Argo Rollouts