Simplifying Cloud Native app testing across environments with Dapr and Microcks

for more info visit: https://www.salaboy.com

Transcript

1. SIMPLIFYING CLOUD-NATIVE APP TESTING ACROSS ENVIRONMENTS Mauricio Salatino - @salaboy Laurent Broudoux - @lbroudoux
2. Demo #1
3. Cloud Native Pizza Store https://github.com/salaboy/pizza
4. Placing an order pseudo code
5. Cloud Native Pizza Store (with infra)
6. Mauricio Salatino @salaboy • Ecosystem Engineer @ Diagrid working on Dapr • An Open Source addict and contributor Who we are!
7. Laurent Broudoux @lbroudoux • A Cloud-native guy - from Financial Services to OSS/Cloud vendors • Loves everything distributed! #API, #OpenAPISpec, #AsyncAPISpec, #Kubernetes, #Integration, #Camel • An Open Source addict and contributor • Works at Postman in Open Technologies Dep. • Dad, Gardener & Pastries lover 🍰 🎂 🧁 🍩 🥧 🍓 Who we are!
8. What we need for an Enterprise Application
9. • Distributed Application Runtime • CNCF Graduated Project • Multi language SDKs • Case Study https://www.cncf.io/case-studies/datagalaxy/
10. Dapr (https://dapr.io)
11. What we used for this demo
12. Let’s start with the basics
13. Service-to-Service Invocation (Store to Kitchen and Delivery)
14. Cloud Native Pizza Store (with Dapr)
15. PubSub from Kitchen and Delivery to Store Service Events
16. Cloud Native Pizza Store (with Dapr)
17. Cloud Native Pizza Store (with Dapr PubSub)
18. Part #1 - Distributed app challenges • #1 Managing cross-cutting concerns (security, resilience, observability, …) • #2 Managing cross-environment concerns (internal vs managed services, AWS vs GCP vs Azure) • Dapr to the rescue!
19. Oops… I added a new service dependency…
20. DevEx & DISTRIBUTED APPS - WHAT A BURDEN! $ mvn spring-boot:run $ ./gradlew bootRun $ npm run dev $ ./mvnw quarkus:dev $ make run $ docker compose up -d $ go run main.go $ ./kafka-server-start.sh $ postgres -D ~/pg-data
21. DevEx & DISTRIBUTED APPS - WHAT THE HELL! The Developerʼs Laptop Service A Service B Service C Database A Database B Database C OK BUT … LOTS OF COMPLEXITY LOTS OF POWER / ENERGY LOTS OF MONEY!
22. DevEx & DISTRIBUTED APPS - TAKING SHORTCUTS • Avoid Testing Integrations • Use mock frameworks only • In-VM vs Network • Rely on your own vision
23. TESTCONTAINERS to the RESCUE!
24. FROM DISTRIBUTED APPS to PROMISES to … Services & APIs Promises Contracts
25. https://microcks.io
26. INGEST PROVIDE MOCKS WELL-KNOWN ARTIFACTS 🧠 SMART MOCKS API Consumer and others MICROCKS Turn Artifacts into light & smart simulations
27. INGEST WELL-KNOWN ARTIFACTS AUTOMATIC TEST SUITE and others VALIDATE REAL API API v1.2 API v1.1 API v1.0 API Provider MICROCKS Turn Artifacts into Contract Testing Suite
28. Docker/AtomicJar partners with software vendors to maintain and certify official modules. Java, NodeJS, Go, …
29. Demo #2
30. Contract Testing in Action! • Start your Application • Include a library like RestAssured • Call API endpoint and check: ◦ Response status code ◦ Response content-type ◦ JSON evaluation ◦ JSON response is valid on all paths ◦ Loop on next call… Typical approach
31. Contract Testing in Action! • Include Microcks and OpenAPI • Launch a Microcks Test • AssertSuccess Microcks approach ✨ : managed by Microcks Testcontainers tests • For all operations and examples: ◦ Act as an API consumer ◦ Validate OpenAPI schema ✨
32. Contract Testing in Action for Async! 😱 • Include a mocking library • Prepare Mock object for DaprMessagingTemplate.send() • Check that messagingTemplate.send() has been called • Tear-down all resources Typical “lightˮ approach
33. Contract Testing in Action for Async! 😱 Microcks approach ✨ : managed by Microcks Testcontainers tests • Include Microcks and AsyncAPI • • Launch a Microcks Test • Invoke KitchenService.emitEvent() • Start a Kafka broker ✨ • Check that messagingTemplate.send() has been called • + that message is sent over the wire • + that message is on correct Kafka topic • + that message is conformant to schema • Tear-down all resources ✨
34. Part #2 - Distributed / DX challenges • #3 Managing more dependencies (complex setup, cost, slow, …) • #4 Testability? Confidence? (integration tests, no breaking changes) • Microcks to the rescue!
35. Building resiliency into our applications
36. Building resiliency into our applications
37. Building resiliency into our applications
38. Dapr Workflows for Durable Executions
39. Workflow as Code (Workflow)
40. Workflow as Code (Activity)
41. Jnation 2025: Durable Executions in Serverless Architectures: Cloudflare, SpringBoot & Dapr by Nele Uhlemann, Mauricio Salatino Workflow Runtime
42. Process Order Workflow • Execute each Activity until it's done • Retry in case of failure • Continue from last activity executed if the application crashes
43. Order from Prompt • Reuse workflows and activities • Integrates with your favourite tools such as Spring AI
44. Demo #3
45. Running Ollama locally 2m 40s on 1Gb internet connection!
46. Building OpenAI mocks with Microcks! OpenAPI (Primary) Collections Import Import API / Service metadata operations examples examples Test suite APIExamples {har} 🧠 LLM (Secondary Artifacts)
47. A simple OpenAI completion mock definition
48. Demo #4
49. A new Testing Nightmare? 🤯 • Prepare an Event object • Invoke PizzaStore.receiveEvents() handling function • Include a mocking library • Prepare Mock object for RestTemplate.put() • Check that restTemplate.put() has been called • Tear-down all resources Typical “lightˮ approach
50. A new Testing Nightmare? 🤯 • Include Microcks and AsyncAPI • • • • • Start a workflow instance & check that DeliveryService is called at some point • Tear-down all resources Microcks approach ✨ : managed by Microcks Testcontainers tests • Start a Kafka broker • Publish mock messages on Kafka • Check that PizzaStore.receiveEvents() is connected and can read message • Check that consumer deserialization is ok ✨ ✨
51. Wrap-up
52. ALL TOGETHER
53. Inner or Outer Loops: Choice is yours! Setup Code Build Test Push Review Build Conform- ance Security Deploy Inner loop Outer loop Pull/Merge Request Developer Team Microcks Microcks Microcks
54. KEY TAKEAWAYS 🤔 • Building distributed applications is hard… • Making these apps portable across environment is possible using APIs, check out the Dapr SDK! • Still, the Developer Experience can be a pain… • Microcks makes mocks easy & contracts finally useful! for a better Developer Experience, for Java, .Net, etc.
55. SIMPLIFYING CLOUD-NATIVE APP TESTING ACROSS ENVIRONMENTS Mauricio Salatino - @salaboy Laurent Broudoux - @lbroudoux THANK YOU! Feedback is a gift!
56. Demo #1 🪄
57. Wrap-up 🏁
60. Agenda • Distributed Applications 20 mins) ◦ The Cloud Native Pizza App Demo #1 ◦ Sync and Async interactions Small Dapr intro) ◦ Testing and Mocking Sync and Async Demo #2 • Building resilient applications 25 mins) ◦ Dapr → Durable executions and Workflows (inner loop) Demo #3 → runs the three services) ◦ Testing with real complex examples OpenAI and Dapr Workflows Demo #4 → mocks everywhere .. only store service) • Takeaways
61. PLACEHOLDER (Part #2) • Building resilient and AI-infused applications challenges • #1 Control of durable executions • Dapr Workflows to the rescue! • #2 Speed & cost when running locally • Microcks to the rescue!
62. PLACEHOLDER (Laurent) • Distributed applications challenges • #3 More dependencies ⇒ DX + laptop setup • #4 How to test things efficiently? How to guarantee your consumers youʼre not introducing breaking changes? • Microcks to the rescue!