Continuous delivering serverless applications in the pipeline: indispensable advice from the trenches

Failure is not the opposite of success, it’s an essential part of it. Failing in a controlled and tested environment helps us to learn and improve; a core principle of continuous delivery. Serverless technology gives us the power to fail faster, and more often. Since there are no servers to manage, you can create isolated, production-like environments much faster. Because of this, it will be easier to create and run pipelines which will run faster and more stable. I will describe the challenges we faced in building serverless applications in a continuous delivery pipeline.

Transcript

1. Continuous delivering serverless applications in the pipeline Kenny Baas AMSTERDAM | MAY 8-9, 2018
2. Kenny Baas Software Consultant - Event Stormer Domain Driven Design Behaviour Driven Development Continuous Delivery @kenny_baas Baasie.com blog.xebia.com/author/kbaas/
3. Dave Farley - Continuous Delivery Feedback Quickly Timely Reliable Customer
4. Dave Farley - Continuous Delivery Customer
6. Software Development is about learning, working code is a side effect - Alberto Brandolini
9. You are either successful at what you are trying to attempt, or you will learn from it.
13. Releasing software is too often an art; it should be an engineering discipline - Dave Farley
14. • Version Control • Automated Configuration • Automated Infrastructure • Deployment Pipeline • Trunk Based Development • Continuous Integration • Evolutionary Architecture Steve Smith - Measuring Continuous Delivery • Database migration • Test Driven Development • Acceptance Test & Smoke test • Exploratory Testing • Dynamic Test Data • Monitoring and Alerting • Canary Deployments Continuous Delivery Technology Changes
17. Dave Farley - Continuous Delivery Build 1. Compile 2. Unit Test 3. Analyse Acceptance Test (Production Like) 1. Configure Environment 2. Deploy Release Candidate 3. Smoke Test/Health Check 4. Acceptance Test Staging Production Release Candidate
22. https://web.devopstopologies.com/
23. #NoOps #OpsLess
24. https://martinfowler.com/bliki/Serverless.html
34. Demo
36. crucial discoveries always emerge during the design/implementation effort. - Eric Evans
38. Recap → Try{} Catch{} → Learn how to deal with failure, and learn from it. → Keep the feedback loop short. → Code, Infra, Pipeline all as Code in your repo and Release Candidate → Discover the bounded context for your pipeline with Event Storming/DDD → Easy setup, but be careful at your configuration!! → Pick your tool for the job. → Incorporate Acceptance Testing from the beginning https://gitlab.com/Baasie/continuous-delivery-faas/
39. Q&A