Having different deployment strategies is essential to ensure that the new version of the software is delivered to users in an efficient and robust way. …

Build API with Spring-Data-Rest and create corresponding OpenFeign client — In this article, we will introduce a convenient library, Spring-Data-Rest to eliminate your meaningless code in a Spring Boot Application. As we all know, a DB-access API will be implemented as a Controller-Service-Repository stack in a Spring Boot application coding convention. Often, there is no business logic in controllers and…

As we mentioned in this article before, centralizing all the needed knowledge in one developer portal is a big improvement in daily working experience and convenience. But we face a CORS problem when sending requests by an embedded swagger page in the API definition. …

A developer portal is designed to enhance the developer experience by uniting all the essential knowledge required for development, maintenance, monitoring, and more, into a single platform. Backstage fulfills this objective through its core features, which include: Software catalog: This feature allows users to define relationships between systems, components, and…

Achieving dead letter queue, blocking and non-blocking retry mechanisms by using RetryableTopic annotation — In the previous article I shared before, I didn’t show how to implement the error handling in the Kafka consumer for our spring boot application. Since the missing piece is so essential, here I wrote a new article to show how to do the following stuff: Blocking retry Do retry when…

Run End to End Test in PR check by GitHub Action — This article will cover the first step for achieving True Continuous Delivery: How to gain more confidence when merging code into the main branch. The answer is always more tests, more specifically, more automated tests. …

A Step-by-Step Guide to Designing Event Sourcing and CQRS Systems — After introducing what benefit can be brought into the system in my last article. I decided to build a Proof of Concept (POC) project with Event Sourcing and CQRS architecture. The first thing is to confirm the project's scope, and there is an excellent theory, Event Storming to do so. …

Building a common module for other modules, projects, or even teams to use is happening repeatedly during our software engineer job. So, how to make it easy to use, and even as a plug-and-work plugin is one of the key points to promote it. In this article, we will introduce…

延續上一篇的故事，我們假設銷售中心如同我們提供服務的系統，目前有了基礎的門口接待人員 (Load balancer)、銷售人員 (server)、銷控人員 (DB)、圖檔室 (File system) 等元件，足以應付一般大部分的狀況。但假如這是一棟上千上萬戶的超大型社區，若還是保持原本的配置，那麼絕對就會發生一些客戶在現場等很久的客訴事件。等什麼呢？可能是銷售人員進去圖檔室拿資料，銷控人員忙著查詢不同的銷售資料太久 (DB/File system query/download too long) 或是出價後要聯絡建商卻打不通等情形，導致銷售人員服務一個客戶花的時間拉長，搞不好還有在外面排隊的客戶無法進來。這些實際上也是系統常常會遇到的困境，而解決方法概念也與系統設計出乎意料的相似，這邊就繼續來以房屋銷售中心的例子來講解系統設計時所用來優化效能的技術與概念。 基本篇 監控調度（SRE）篇 (to be done)