Using Redis in SpringBoot

• docs
  • https://docs.spring.io/spring-data/data-redis/docs/2.5.4/reference/html/#introduction
  • https://dgempiuc.medium.com/spring-websocket-and-redis-pub-sub-a02af0dabddb
  • https://www.baeldung.com/spring-data-redis-pub-sub

High level

1. ChatController → RedisPublisher → Redis
2. Redis → All Servers (RedisSubscriber)
3. RedisSubscriber → SimpMessagingTemplate → WebSocket Clients

Flow example

1. The Trigger (Controller)
   • File: ChatController.java Location: Server 1
   • User A sends a WebSocket message to /pub/.... The ChatController receives it. Instead of calling SimpMessagingTemplate (which only talks to local users), it calls your RedisPublisher.

// Inside ChatController
redisPublisher.publish(destination, chatDto);

2. The Output (Publisher)
   • File: RedisPublisher.java Location: Server 1
   • This class uses RedisTemplate (a Spring Bean) to serialize your Java object (the DTO) into JSON and push it out of the Java application and into the Redis database.

// Inside RedisPublisher
// "chatTopic" is a specific channel name in Redis, like "chatroom-channel"
redisTemplate.convertAndSend(chatTopic.getTopic(), payload); 

• At this exact moment, the message has left Server 1.

3. The Relay (Redis)
   • Location: Redis Instance (AWS ElasticCache or Docker)
   • Redis receives the JSON payload on the “chatroom-channel”. It immediately pushes that JSON to every server that is currently “listening” to that channel.
     • It pushes to Server 1.
     • It pushes to Server 2.
4. The Listener (Container & Adapter)
   • File: RedisConfig.java (The Beans) Location: Server 1 AND Server 2
   • This is what the Beans in RedisConfig are actually doing:
     1. RedisMessageListenerContainer
        • a background thread running on your server.
        • holds an open connection to Redis and waits for data → When Redis pushes the JSON (from Step 3), this container catches it
     2. MessageListenerAdapter
        • The Container hands the raw data to this Adapter. The Adapter knows specifically which method in your code to call (e.g., "onMessage").
5. The Processor (Subscriber)
   • File: RedisSubscriber.java Location: Server 1 AND Server 2
   • The Adapter calls the onMessage method in this class. The message is now back inside your Java application.
     1. It deserializes the JSON back into a Java Object (RedisMessagePayload).
     2. Now it uses SimpMessagingTemplate.

// Inside RedisSubscriber
// contentId is extracted from the payload
messagingTemplate.convertAndSend(payload.getDestination(), payload.getContent());

6. The Final Push (WebSocket)
   • Location: Server 1 AND Server 2
   • SimpMessagingTemplate looks at the local WebSocket session registry.
     • Server 1: Sees User A and User B are connected. It sends the message to them.
     • Server 2: Sees User C is connected. It sends the message to them.