WebSocket Architecture

Overview

Lichess handles real-time communication through a separate WebSocket server called lila-ws, which communicates with the main Scala application (lila) via Redis pub/sub. This architecture enables horizontal scaling and handles millions of concurrent WebSocket connections.

Architecture Diagram

┌─────────────┐
│   Browser   │
└──────┬──────┘
       │ WebSocket
       ↓
┌─────────────┐
│    nginx    │  (WebSocket proxy)
└──────┬──────┘
       ↓
┌─────────────┐
│   lila-ws   │  (WebSocket server)
│  (Scala 3)  │  - Manages connections
└──────┬──────┘  - Handles real-time events
       │
       │ Redis Pub/Sub
       ↓
┌─────────────┐
│    Redis    │  (Message broker)
└──────┬──────┘
       │
       ↓
┌─────────────┐
│    lila     │  (Main application)
│  (Scala 3)  │  - Game logic
└─────────────┘  - HTTP requests

Why Separate WebSocket Server?

Benefits

Horizontal scaling: Run multiple lila-ws instances behind load balancer Isolation: WebSocket connections don’t consume resources from main app Resilience: Main app restarts don’t drop WebSocket connections Performance: Optimized for connection handling vs. business logic Language flexibility: Could rewrite in different language if needed

Trade-offs

Complexity: Additional service to deploy and monitor Latency: Extra hop through Redis adds ~1-5ms Consistency: Must handle Redis connection failures gracefully

lila-ws Server

Technology Stack

Language: Scala 3
HTTP: Netty-based HTTP server
Concurrency: Akka actors for connection management
Redis: Lettuce client for pub/sub

GitHub: lichess-org/lila-ws

Core Responsibilities

Connection Management

Maintain WebSocket connections for:Game rooms: Players and spectators watching a game

/watch/{gameId} - Spectate game
/play/{gameId} - Play in game

Tournament rooms: Tournament participants and viewers

/tournament/{tournamentId}

Study rooms: Collaborative analysis boards

/study/{studyId}

User notifications: Personal event streams

/user/{userId}

Lobby: Players seeking games

/lobby

Site-wide: Global announcements

/site

Message Routing

lila-ws routes messages between clients and lila:Client → lila:

// Client sends move
{"t": "move", "d": {"from": "e2", "to": "e4"}}

// lila-ws forwards via Redis to lila
redis.publish("site-in", payload)

lila → Client:

// lila publishes game update
redis.publish("game:abc123", moveData)

// lila-ws receives from Redis, sends to clients
socket.send(moveData)

Presence Tracking

lila-ws tracks online users:

// Maintains set of online user IDs
private val onlineUsers: mutable.Set[UserId] = mutable.Set.empty

// Publish online count periodically
scheduler.scheduleWithFixedDelay(1.second, 1.second):
  redis.publish("users-online", onlineUsers.size)

Used for:

“Online” indicators on profiles
Friend online notifications
Activity statistics

Connection Lifecycle

// Simplified connection flow

// 1. Client connects
GET /watch/abc123 HTTP/1.1
Upgrade: websocket

// 2. lila-ws accepts connection
val socket = new Socket(connection, gameId)
sockets.add(socket)

// 3. Subscribe to Redis channel
redis.subscribe(s"game:${gameId}")

// 4. Send initial state
socket.send(JsonData(currentGameState))

// 5. Handle messages
socket.onMessage: msg =>
  parseMessage(msg) match
    case Move(from, to) => 
      redis.publish("site-in", MoveData(gameId, from, to))
    case Chat(text) => 
      redis.publish("site-in", ChatData(gameId, text))

// 6. Client disconnects
socket.onClose:
  redis.unsubscribe(s"game:${gameId}")
  sockets.remove(socket)

Redis Communication

Redis Channels

Lichess uses Redis pub/sub with predictable channel patterns: Site input (clients → lila):

site-in    # All client actions

Game channels (lila → clients):

game:{gameId}           # Game-specific events
tournament:{tournamentId}
study:{studyId}
user:{userId}
lobby
site                    # Global announcements

Message Protocol

Messages use JSON with type discriminator:

// Client to server
interface ClientMessage {
  t: string;    // Message type: "move", "chat", "draw", etc.
  d: unknown;   // Message data
}

// Server to client
interface ServerMessage {
  t: string;    // Event type: "move", "crowd", "end", etc.
  d: unknown;   // Event data
}

Example: Game Move

1. Client sends move:

{"t": "move", "d": {"u": "e2e4", "b": 1}}

2. lila-ws publishes to Redis:

redis.publish("site-in", {
  "gameId": "abc123",
  "userId": "player1",
  "type": "move",
  "move": "e2e4"
})

3. lila validates and processes move 4. lila publishes result to Redis:

redis.publish("game:abc123", {
  "t": "move",
  "d": {
    "ply": 1,
    "uci": "e2e4",
    "san": "e4",
    "fen": "rnbqkbnr/pppppppp/8/8/4P3/8/PPPP1PPP/RNBQKBNR",
    "clock": {"white": 600, "black": 600}
  }
})

5. lila-ws broadcasts to all game watchers

lila Integration

Socket Module

The main lila application communicates with lila-ws through the socket module:

// modules/socket/src/main/Env.scala
final class Env(
  appConfig: Configuration,
  shutdown: CoordinatedShutdown
)(using Executor, Scheduler):
  
  private val redisClient = RedisClient.create(
    RedisURI.create(appConfig.get[String]("socket.redis.uri"))
  )
  
  val remoteSocket: RemoteSocket = wire[RemoteSocket]
  
  // Subscribe to incoming messages from lila-ws
  remoteSocket.subscribe(
    "site-in",
    RemoteSocket.Protocol.In.baseReader
  )(remoteSocket.baseHandler)
  
  // Track online users
  val onlineIds = OnlineIds(() => remoteSocket.onlineUserIds.get)

RemoteSocket API

final class RemoteSocket(
  redisClient: RedisClient,
  lifecycle: play.api.inject.ApplicationLifecycle
)(using Executor):
  
  // Publish message to specific channel
  def publish(channel: String, data: JsonData): Unit =
    redis.publish(channel, Json.stringify(data))
  
  // Send to all clients in a game
  def sendToGame(gameId: GameId, data: JsonData): Unit =
    publish(s"game:$gameId", data)
  
  // Send to specific user
  def sendToUser(userId: UserId, data: JsonData): Unit =
    publish(s"user:$userId", data)
  
  // Broadcast to all connected clients
  def broadcast(data: JsonData): Unit =
    publish("site", data)
  
  // Get set of online user IDs
  def onlineUserIds: Future[Set[UserId]] =
    // lila-ws periodically publishes online user count
    Future.successful(cachedOnlineUsers)

Round Module Integration

The round module uses sockets for live game updates:

// modules/round/src/main/Env.scala
final class Env(
  socketKit: lila.core.socket.ParallelSocketKit,
  // ...
):
  
  // Notify players and spectators of move
  def broadcastMove(game: Game, move: Move, fen: Fen): Unit =
    socketKit.sendToGame(game.id, Json.obj(
      "t" -> "move",
      "d" -> Json.obj(
        "ply" -> game.ply,
        "uci" -> move.uci,
        "san" -> move.san,
        "fen" -> fen,
        "clock" -> game.clock.map(clockJson)
      )
    ))

WebSocket Message Types

Game Room Messages

Client → Server:

move - Player makes a move
rematch - Player offers rematch
takeback - Request takeback
draw - Offer/accept draw
resign - Resign game
abort - Abort game
moretime - Give opponent more time
palantir - Premium user sees opponent’s time usage

Server → Client:

move - Move made (includes fen, clock, etc.)
end - Game ended (winner, reason)
crowd - Spectator count changed
clock - Clock sync
gone - Player disconnected
goneIn - Player will be considered gone soon
premove - Premove acknowledged

Tournament Room Messages

Server → Client:

reload - Tournament data changed (refetch)
redirect - Player should join game
standing - Leaderboard update

Lobby Messages

Client → Server:

join - Join lobby pool
cancel - Leave lobby pool

Server → Client:

redirect - Game found, join it
pools - Pool counts updated

Scalability

Horizontal Scaling

Multiple lila-ws instances can run simultaneously:

┌──────────┐
│  nginx   │  (Load balancer)
└────┬─────┘
     ├──────────┬──────────┐
     ↓          ↓          ↓
┌─────────┐┌─────────┐┌─────────┐
│ lila-ws ││ lila-ws ││ lila-ws │
│  node1  ││  node2  ││  node3  │
└────┬────┘└────┬────┘└────┬────┘
     └──────────┴──────────┘
              ↓
         ┌────────┐
         │ Redis  │  (Shared pub/sub)
         └────────┘

Sticky sessions: Nginx routes user to same lila-ws node (optional) Failover: If node dies, clients reconnect to another node

Connection Limits

Each lila-ws instance handles:

100,000+ concurrent WebSocket connections
10,000+ messages per second

Production deployment typically runs 3-5 lila-ws instances.

Redis Performance

Redis pub/sub can handle:

100,000+ messages/second
Sub-millisecond latency

Redis Cluster used for high availability:

3+ Redis nodes with replication
Automatic failover
Sentinel for monitoring

Client-Side WebSocket

TypeScript Socket Client

Frontend uses a custom WebSocket wrapper:

// ui/analyse/src/socket.ts
import { make as makeSocket } from 'lib/socket';

export function make(send: SocketSend, ctrl: AnalyseCtrl): Socket {
  const socket = makeSocket(
    `/watch/${ctrl.data.game.id}`,
    ctrl.socketReceive
  );
  
  return {
    send: socket.send,
    receive: socket.receive,
    destroy: socket.destroy
  };
}

Socket Library

// ui/lib/src/socket.ts
export function make(url: string, onMessage: (data: any) => void) {
  let ws: WebSocket;
  let pinging = false;
  
  const connect = () => {
    ws = new WebSocket(`wss://${location.host}${url}`);
    
    ws.onopen = () => {
      console.log('Socket connected');
      startPinging();
    };
    
    ws.onmessage = (e) => {
      const data = JSON.parse(e.data);
      onMessage(data);
    };
    
    ws.onclose = () => {
      console.log('Socket closed, reconnecting...');
      stopPinging();
      setTimeout(connect, 2000);  // Reconnect after 2s
    };
  };
  
  const send = (type: string, data: any) => {
    if (ws.readyState === WebSocket.OPEN) {
      ws.send(JSON.stringify({ t: type, d: data }));
    }
  };
  
  // Keep-alive pings
  const startPinging = () => {
    pinging = true;
    const ping = () => {
      if (pinging && ws.readyState === WebSocket.OPEN) {
        ws.send(JSON.stringify({ t: 'p' }));  // Ping
        setTimeout(ping, 2000);
      }
    };
    ping();
  };
  
  connect();
  return { send, destroy: () => ws.close() };
}

Reconnection Strategy

Automatic reconnect: Exponential backoff (2s, 4s, 8s, …)
State recovery: Server sends full state on reconnect
Buffering: Queue messages while disconnected

Monitoring

Metrics

Key metrics tracked:

Connection count: Active WebSocket connections per node
Message rate: Messages/second in/out
Redis lag: Time between publish and receive
Reconnection rate: Clients reconnecting (indicates issues)
Error rate: Failed message parsing, Redis timeouts

Health Checks

// lila-ws health endpoint
GET /health

{
  "connections": 42315,
  "redis": "connected",
  "uptime": 86400
}

Get Started

Core Features

Development

Architecture

Contributing

WebSocket Architecture

Overview

Architecture Diagram

Why Separate WebSocket Server?

Benefits

Trade-offs

lila-ws Server

Technology Stack

Core Responsibilities

Connection Lifecycle

Redis Communication

Redis Channels

Message Protocol

Example: Game Move

lila Integration

Socket Module

RemoteSocket API

Round Module Integration

WebSocket Message Types

Game Room Messages

Tournament Room Messages

Lobby Messages

Scalability

Horizontal Scaling

Connection Limits

Redis Performance

Client-Side WebSocket

TypeScript Socket Client

Socket Library

Reconnection Strategy

Monitoring

Metrics

Health Checks

See Also

Get Started

Core Features

Development

Architecture

Contributing

Documentation Index

​Overview

​Architecture Diagram

​Why Separate WebSocket Server?

​Benefits

​Trade-offs

​lila-ws Server

​Technology Stack

​Core Responsibilities

​Connection Lifecycle

​Redis Communication

​Redis Channels

​Message Protocol

​Example: Game Move

​lila Integration

​Socket Module

​RemoteSocket API

​Round Module Integration

​WebSocket Message Types

​Game Room Messages

​Tournament Room Messages

​Lobby Messages

​Scalability

​Horizontal Scaling

​Connection Limits

​Redis Performance

​Client-Side WebSocket

​TypeScript Socket Client

​Socket Library

​Reconnection Strategy

​Monitoring

​Metrics

​Health Checks

​See Also

Overview

Architecture Diagram

Why Separate WebSocket Server?

Benefits

Trade-offs

lila-ws Server

Technology Stack

Core Responsibilities

Connection Lifecycle

Redis Communication

Redis Channels

Message Protocol

Example: Game Move

lila Integration

Socket Module

RemoteSocket API

Round Module Integration

WebSocket Message Types

Game Room Messages

Tournament Room Messages

Lobby Messages

Scalability

Horizontal Scaling

Connection Limits

Redis Performance

Client-Side WebSocket

TypeScript Socket Client

Socket Library

Reconnection Strategy

Monitoring

Metrics

Health Checks

See Also