aenesgur/microservice-saga-outbox-inbox-pattern

This repo contains Saga, Outbox, Inbox and Distributed Lock Patterns with a sample Spring Boot application.

Microservice Application with Saga, Outbox, Inbox, Distributed Locking Design Patterns

Microservice boilerplate application developed with Outbox/Inbox pattern and Distributed Locking

Technologies

• Spring Boot
• Spring Data
• Apache Kafka
• Redis
• Docker
• Postgresql

What is Inbox/Outbox pattern?

The Inbox-Outbox pattern is a messaging pattern used in distributed systems to improve the performance and reliability of communication between components.

In this pattern, each component has two message queues: an inbox and an outbox. The inbox receives messages from other components, while the outbox holds messages that the component needs to send to other components.

The component processes messages from its inbox in a separate thread, and once it completes processing a message, it places the response message in its outbox. The outbox messages are then sent by a separate thread that continuously polls the outbox queue for new messages and sends them to their destinations.

This pattern can help improve performance and reliability by decoupling the sending and receiving of messages, reducing contention and preventing message loss. It also enables components to operate asynchronously, allowing them to continue processing messages from their inbox while messages from their outbox are being sent.

What is Distributed Locking?

Distributed locking is a technique used in distributed computing systems to ensure that multiple processes or threads don't interfere with each other when accessing shared resources, such as data or files. The basic idea behind distributed locking is to use a shared lock that multiple processes or threads can acquire and release as needed.

Distributed locking is important in distributed systems because it helps prevent conflicts and inconsistencies that can arise when multiple processes or threads try to access the same shared resource simultaneously. By using a distributed lock, each process or thread can take turns accessing the shared resource in a coordinated and orderly way, which helps maintain system stability and reliability.

System Diagram

Distributed Lock Diagram

Run

First: mvn clean install

Then: docker compose up -d

Test

Test for payment-fail-topic:

curl --location --request POST 'localhost:4041/api/orders' \ --header 'Content-Type: application/json' \ --data-raw '{ "totalPrice": 12, "cardNo":"27411111222233334444", "userId": 45623, "orderNote":"lorem ipsum" }'

Test for payment-complete-topic:

curl --location --request POST 'localhost:4041/api/orders' \ --header 'Content-Type: application/json' \ --data-raw '{ "totalPrice": 12, "cardNo":"77411111222233334444", "userId": 45623, "orderNote":"lorem ipsum" }'