Managing projects with both a front and back end. Read More »

Running Light-4J Application as Linux Service Read More »

Running Light-4J Application as Windows Service Read More »

There are some dependent services that need to be started in order to do the development of light-eventuate-4j, light-tram-4j and light-saga-4j. On Linux and Mac, we usually start all of them from a docker-compose as it is described in this tutorial. For Windows users, AFAIK docker-compose doesn’t work. The following is a manual process to install Kafka, Zookeeper and CDC server for tram and CDC server for eventuate. In production, at least Kafka and Zookeeper will be installed as a cluster on Linux bear metal or VMs. Read More »

After you complete your service development, you need to consider how to deploy your service to the production with security in mind. In this section, we will discuss some of the security measurements that help you to prevent attacks from the Internet when you expose your service to the public. Services built on top of light-4j are inherently more secure than most platforms as we have a lot of middleware handlers to address cross-cutting concerns and a significant number of them are for security. Read More »

High throughput and low latency are principles of Light. When it comes to service to service communication, we recommend using client side service discovery and load balancing. light-4j has a client module named Http Client that can be embedded into any Java application to handle the communication to services built in Light. It handles HTTP 2.0 connection and OAuth 2.0 provider communication as well as traceability in the service to service call tree. Read More »

Java Heap Memory Limit in Docker Java Heap Memory Limit in Docker Note: This is the issue with JDK 8 for 1.6.x and 1.5.x releases. If you are using JDK11 with the 2.0.x release, you don’t need to worry about it. The issue is that if the Xmx option is not defined explicitly, then JVM uses 1/4th of all memory available for the host OS due to a default internal garbage collection (GC) ergonomic algorithm. Read More »

Although you can deploy the microservices built on top of Light to a data center with java -jar or with Docker, the ultimate deployment platform would be a Kubernetes cluster. Before deploying a service to a Kubernetes cluster, you need to create secrets with all the externalized configuration files. These files usually are copied to the config folder in each project in llight-config-test or light-config-prod. For a detailed directory structure, please refer to petstore. Read More »