惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

C
Check Point Blog
罗磊的独立博客
量子位
Microsoft Azure Blog
Microsoft Azure Blog
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
J
Java Code Geeks
M
MIT News - Artificial intelligence
月光博客
月光博客
IT之家
IT之家
D
DataBreaches.Net
A
About on SuperTechFans
博客园 - 三生石上(FineUI控件)
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
Last Week in AI
Last Week in AI
D
Docker
The GitHub Blog
The GitHub Blog
B
Blog
V
Visual Studio Blog
博客园 - Franky
N
Netflix TechBlog - Medium
博客园 - 【当耐特】
Martin Fowler
Martin Fowler
博客园 - 聂微东
U
Unit 42

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant
The Stale Closure Bug That Haunted My Chat App
Nikhil Sharma · 2026-06-14 · via DEV Community

Building a real-time chat app sounds straightforward until React's rendering model meets WebSockets. In Relay, I built what I thought was a solid real-time integration. Everything seemed to work perfectly—until I started switching between conversations.

What I Built

In Relay, I wanted a clean separation of concerns, so I extracted my WebSocket logic into a custom hook called useWebSocket. It takes an onMessage callback so my Chat UI can decide how to handle incoming messages:

// Chat.tsx
const { sendMessage } = useWebSocket((msg) => {
  // If the message is from the person I'm actively chatting with, add it to the window
  if (selectedUser && msg.senderId === selectedUser.id) {
    addMessage(msg);
    return;
  }

  // Otherwise, show a toast notification
  toast(`${msg.sender.name}: ${msg.content}`);
});

Inside useWebSocket.ts, I wired up the actual WebSocket connection inside a useEffect:

// useWebSocket.ts (The broken version)
export function useWebSocket(onMessage: (msg: Message) => void) {
  const { user } = useAuth();
  useEffect(() => {
    const socket = new WebSocket(WS_URL);

    socket.onmessage = (e) => {
      const msg = JSON.parse(e.data);
      if (msg.type === "receive_message") {
        onMessage(msg.payload);
      }
    };
    return () => socket.close();
  }, [user]); // Only reconnect if the authenticated user changes
}

What Broke

At first glance, this code looks entirely reasonable. I'd open Relay, click on Alice's profile, and we'd start chatting. It worked flawlessly.

Then, I'd switch my active chat to Bob. Bob would send me a message, but instead of the message appearing in Bob's chat window, a toast notification would pop up. Even worse, if Alice sent me a message while I was looking at Bob's chat, her message silently injected itself directly into Bob's chat window!

I was literally watching the messages of one user show up in another user's chat after I switched.

Why It Broke (The Mental Model)

The culprit was a classic React gotcha: the stale closure.

When the useWebSocket hook ran for the first time, the useEffect established the WebSocket connection. The socket.onmessage event listener was created, and it permanently "captured" the onMessage function from that specific initial render.

In that first render, selectedUser was Alice.

When I clicked on Bob in the sidebar, React re-rendered my Chat component. A brand new onMessage function was created where selectedUser was Bob. However, because my useEffect dependency array only included user, the effect didn't re-run. The WebSocket stayed connected—which is what we want.

But this meant the socket.onmessage listener was still holding onto the very first onMessage function it captured—the one where selectedUser is permanently frozen as Alice.

If I simply added onMessage to the useEffect dependency array to fix the stale state, the WebSocket would brutally disconnect and reconnect every single time I clicked a new user, typed a character, or triggered a re-render. That's terrible UX and a massive performance killer.

The useRef Fix

I needed a way for the WebSocket listener to always access the freshest version of the onMessage callback, without actually triggering the useEffect to tear down the socket.

Enter useRef.

Unlike state, updating a ref doesn't trigger a re-render. More importantly, a ref is a mutable object that persists across renders. By continuously storing the latest callback in a ref, the WebSocket listener can just peek into ref.current and always find the most up-to-date function.

Here is the exact code that fixed the bug in Relay:

// useWebSocket.ts (The fixed version)
export function useWebSocket(onMessage: (msg: Message) => void) {
  const { user } = useAuth();

  // 1. Create a ref to hold the callback
  const onMessageRef = useRef(onMessage);
  // 2. Silently update the ref every time the callback changes (on every render)
  useEffect(() => {
    onMessageRef.current = onMessage;
  }, [onMessage]);
  useEffect(() => {
    const socket = new WebSocket(WS_URL);

    socket.onmessage = (e) => {
      const msg = JSON.parse(e.data);
      if (msg.type === "receive_message") {
        // 3. Always call the freshest callback stored in the ref!
        onMessageRef.current(msg.payload);
      }
    };
    return () => socket.close();
  }, [user]); 
}

The Takeaway

Whenever you integrate long-lived event listeners (like WebSockets, window events, or setInterval) with React components, you have to be paranoid about what variables are being captured in their closures. If you need access to constantly changing React state inside a permanent event listener, the useRef escape hatch is your best friend.

I'll be back with more next week. Until then, stay consistent!

Sidenote

Relay is now live: https://relay.nikshrma.dev

It's running on a DigitalOcean droplet and has been a great playground for learning WebSockets, authentication, testing, and deployment.

You can also find my portfolio at https://nikshrma.dev.