Testing Reactive Apps
summary
Argues that testing reactive apps requires concrete practices: run tests on production‑like hardware, make timeouts externally configurable, and exercise blue‑green slices and microservice-level deployments. Use property‑based tools like ScalaCheck, inject failures with a Gremlin/failure injector or chaos monkey, and validate SLAs under concurrent load rather than single requests. Leverage ZooKeeper/Kubernetes/Mesos and Docker APIs for node and cluster checks, model errors as domain events, and use Akka TestKit’s multi‑node conductor to simulate partitions and timing nondeterminism.
AI-written summary · about
chapters (26)
- 0:00Intro
- 2:27Who Needs Tests
- 2:53Your Business depends on this
- 3:33I dont like walking around feeling like
- 4:11Reactive Manifesto
- 5:20Testing Frameworks
- 8:08Error vs Failure
- 9:03Whitebox vs Blackbox
- 10:36Test Environment
- 13:29Multiple Slices
- 15:15Sickness Components
- 17:48Asynchronous Testing
- 23:00Scaling
- 24:41Resilience
- 26:25Domain Driven Design
- 28:41ABI Resilience
- 29:35Infrastructure Resilience
- 29:56Network Resilience
- 31:12Cluster Resilience
- 32:32Data Center Resilience
- 33:46Data Center Ops
- 34:25Latency Profiles
- 36:10Summary
- 43:10Message interactions
- 44:46Mocking
- 46:49Daily Observer
related talks
30mHow to push a ReactJS application in production...
17mTesting in a Reactive Way
52mBuilding Reactive Apps
45mReactive Testing Strategies with NgRx
42mReacting to the Future of Application Architecture
57mAdvanced Testing Techniques: Tips from the trenches
40mTop Mistakes When Writing Reactive Applications
48mReactive for the Impatient