review-service-h2

to create microservices faster, you need a backbone. This boilerplate prepared to practise core microservice concepts like:

• cloud-native
  • stateless on K8s
  • state-full on GCP PaaS (not on K8s)
• REST maturity (L2)
• naming conventions
• actuator (/health)
• error-handling
• logging
  • general {Slf4j}
  • message format standardization
  • distributed-tracing {sleuth}
  • monitoring
• api-management
  • message-converter (such as text2json and vice versa)
• security
• profiles (dev, test, prod)
• logical boundaries (separating dbs)
• caching
  • L2 {JCache | Ehcache}
• rest-template {Feign}
• api-gateway {zuul}
• load-balancing {ribbon}
• fallbacks {hystrix}
• config-server {spring}
• service-discovery {eureka}
• event-sourcing
• CQRS (separate read and write queries)

Technology stack

• Requirements (openjdk 11, docker, k8s)
• spring boot 2.4.x, gradle 6.7, Lombok 1.18, log4j 2.x, Ehcache 3.x, junit 5.x

Naming Conventions

• project name for a real service will take place below!
  • intellij top-level: review-service
  • package name: com.backbone.core
  • gradle's project name: review-service
    • jar name: build/libs/review-service-0.0.1-SNAPSHOT.jar
  • deployment name in k8s: review-service
  • url format: /review/1 or /reviews

OOP Architecture

• Review ... as Entity
• ReviewRepository
  • ReviewService (as JPA cache layer or call APIs)
    • ReviewController (as REST APIs)

How To Start

on IDE,

1. mkdir microservice-backbone-boilerplate && cd microservice-backbone-boilerplate then
   • git clone https://github.com/tansudasli/review-service-h2.git && cd review-service-h2
2. to Run the application
   
   • ./gradlew wrapper to download necessary wrappers. Then
   • ./gradlew bootRun to start,
   • ./gradlew bootJar to create jar lib, and run w/ java -jar build/libs/*.jar
3. to test run curl localhost:8080
   • curl localhost:8080/dummy or curl localhost:8080/dummy/name
   • curl localhost:8080/reviews or curl localhost:8080/reviews/10
4. to access h2-db check localhost:8080/h2 w/ conn. jdbc:h2:mem:reviews

on Kubernetes,

1. Create GKE cluster by following in Readme.md.
2. Prepare and deploy docker images to GCP
   • Edit gradle.properties
   • Run ./gradlew dockerTag. Test w/
     • docker images then docker run -d -p 8080:8080 IMAGE_NAME:TAG
   • Run ./gradlew dockerPushGCP
3. Deploy app. to GKE cluster.
   • Run ./deploy.sh to deploy w/ .yaml files
   • to test, run kubectl get services and get EXTERNAL-IP then curl EXTERNAL-IP:8080 to test.

High Level Architecture -

an abstract representation of a micro service

a more realistic scenario w/ data pipelines