WebSockets & Realtime Systems Programming · レッスン

切断と再接続への対応

切断を検出し、自動的に再接続を試みるクライアント側およびサーバー側のロジックを実装します。

「切断と再接続への対応」はCoddyKit上の無料WebSockets & Realtime Systems Programmingレッスンです。これはレッスン1/4です。下記で完全なレッスンを無料で読むことができます。その後、ブラウザ内の組み込みコードエディタと24時間対応のAIチューターでハンズオン演習できます。これはWebSockets & Realtime Systems Programming学習パスの一部であり、ウェブとCoddyKitアプリ全体で進捗が同期されます。 WebSockets & Realtime Systems Programmingコースには全4レッスンが含まれています。

このレッスンの一部はまだ翻訳されておらず、英語で表示されています。

Why Connections Drop

In the real world, network connections aren't always perfect. WebSockets, despite being persistent, can break due to many reasons.

Ignoring these disconnections can lead to unresponsive applications and a poor user experience. This lesson explores how to gracefully handle these interruptions.

Causes of Connection Loss

What makes a WebSocket connection drop?

Network Issues: Wi-Fi drops, internet outages, proxy problems.
Server Restarts: The server application might restart for updates or maintenance.
Client Offline: The user closes their browser, loses power, or puts their device to sleep.
Idle Timeouts: Some proxies or firewalls might close idle connections.

Client Detects Disconnects

On the client side, the browser's native WebSocket API provides events to notify you about connection status changes.

The most important event for detecting a disconnection is the onclose event. It fires when the connection is either gracefully closed by the server or abruptly lost.

Simple Reconnect Attempt

When onclose fires, you can immediately try to reconnect. This simple approach works for temporary glitches but can be problematic.

Try running this basic client-side example:

let ws;
function connect() {
  ws = new WebSocket("ws://localhost:8080");

  ws.onopen = () => console.log("Connected!");
  ws.onmessage = (event) => console.log("Received:", event.data);
  ws.onerror = (error) => console.error("WebSocket Error:", error);

  ws.onclose = () => {
    console.log("Disconnected. Reconnecting...");
    setTimeout(connect, 1000); // Try again in 1 second
  };
}
connect();

Why Simple Retries Fail

Continuously trying to reconnect immediately (e.g., every 1 second) can overwhelm both the client and the server.

Client Resource Drain: Constant connection attempts consume CPU and network resources.
Server Overload: If many clients disconnect and then flood the server with reconnect requests, it can lead to a Denial of Service (DoS) situation.
No Backoff: It doesn't account for prolonged outages.

Smart Reconnection: Backoff

To prevent overloading and provide a better experience, we use reconnection backoff strategies.

A backoff strategy introduces delays between reconnection attempts, and these delays typically increase with each failed attempt. This gives the server time to recover and prevents a "thundering herd" problem.

Exponential Backoff Logic

Exponential backoff is a common and effective strategy. It means the delay between retries increases exponentially.

First retry: 1 second
Second retry: 2 seconds
Third retry: 4 seconds
Fourth retry: 8 seconds

You usually cap the maximum delay to prevent excessively long waits and add some random "jitter" to avoid synchronized retries.

Client with Exponential Backoff

Let's enhance our client-side code to use exponential backoff with a maximum delay and some random jitter. Try running it!

let ws;
let reconnectInterval = 1000; // Start with 1 second
const maxReconnectInterval = 30000; // Max 30 seconds
let reconnectTimer;

function connect() {
  ws = new WebSocket("ws://localhost:8080");

  ws.onopen = () => {
    console.log("Connected!");
    reconnectInterval = 1000; // Reset on successful connection
    clearTimeout(reconnectTimer);
  };
  ws.onmessage = (event) => console.log("Received:", event.data);
  ws.onerror = (error) => console.error("WebSocket Error:", error);

  ws.onclose = () => {
    console.log(`Disconnected. Retrying in ${reconnectInterval / 1000}s...`);
    reconnectTimer = setTimeout(connect, reconnectInterval);
    reconnectInterval = Math.min(reconnectInterval * 2, maxReconnectInterval);
    // Add some random jitter (e.g., +/- 10%) for better distribution
    reconnectInterval += (Math.random() - 0.5) * reconnectInterval * 0.2;
    reconnectInterval = Math.floor(reconnectInterval);
  };
}
connect();

Server-Side Disconnects

The server also needs to know when a client disconnects. This is crucial for cleaning up resources, updating connected user lists, or stopping data streams for that client.

Using a library like ws in Node.js, the WebSocket server emits a 'close' event on the individual client connection object when it disconnects.

Server Cleanup on Disconnect

Here's a basic Node.js WebSocket server example showing how to handle client disconnections and remove them from an active connections list.

const WebSocket = require('ws');
const wss = new WebSocket.Server({ port: 8080 });

let clients = new Set();

wss.on('connection', function connection(ws) {
  clients.add(ws);
  console.log('Client connected. Total:', clients.size);

  ws.on('message', function incoming(message) {
    console.log('received: %s', message);
  });

  ws.on('close', function close() {
    clients.delete(ws);
    console.log('Client disconnected. Total:', clients.size);
  });

  ws.send('Welcome!');
});

console.log('WebSocket server started on port 8080');

Check Your Understanding

Consider a client-side WebSocket application that needs to handle disconnections robustly.

Which of the following are good practices for implementing a reconnection strategy?

Recap: Robust Connections

We've explored how crucial it is to handle disconnections in realtime applications. Both client and server play a role in maintaining connection resilience.

Clients use onclose to detect disconnections.
Exponential backoff with jitter and a max delay prevents server overload during reconnections.
Servers use on('close') to clean up resources when clients leave.

These strategies ensure your realtime apps remain responsive and reliable, even in unstable network conditions.

よくある質問

「切断と再接続への対応」レッスンは無料ですか？

はい。「切断と再接続への対応」の完全なテキストはこのウェブで無料で読めます。インタラクティブに演習し（組み込みコードエディタと24時間対応のAIチューター）、WebSockets & Realtime Systems Programmingコースの残りをアンロックするには、CoddyKit PROにアップグレードしてください。 WebSockets & Realtime Systems Programmingコースには全4レッスンが含まれています。

「切断と再接続への対応」で何を学びますか？

切断を検出し、自動的に再接続を試みるクライアント側およびサーバー側のロジックを実装します。ブラウザで直接実行するハンズオンコードでWebSockets & Realtime Systems Programmingを演習し、24時間対応のAIチューターがレッスンを進める中での質問に答えます。

WebSockets & Realtime Systems Programmingを始めるのに経験は必要ですか？

事前経験は必要ありません。CoddyKitのWebSockets & Realtime Systems Programmingは初級者から上級者向けに構成されているため、ここから始めるか最初から始めて、自分のペースで進むことができます。これはレッスン1/4です。

「切断と再接続への対応」レッスンにはどのくらい時間がかかりますか？

ほとんどのCoddyKitレッスンは約5～10分かかります。各レッスンはコンパクトでインタラクティブなので、着実に進歩し、ウェブとアプリ全体で正確に前回の場所から再開できます。

このWebSockets & Realtime Systems Programmingレッスンでコードを書いて実行できますか？

はい。すべてのWebSockets & Realtime Systems Programmingレッスンに組み込みコードエディタが含まれているため、ブラウザでリアルコードを書いて実行し、即座のAIフィードバックを取得できます。ローカル設定は不要です。

このコースのすべてのレッスン

← WebSockets & Realtime Systems Programmingに戻る