To run multiple HTTP servers concurrently, use the reusePort option in Bun.serve() which shares the same port across multiple processes.
This automatically load balances incoming requests across multiple instances of Bun.
import { serve } from "bun";
const id = = Math.random().toString(36).slice(2);
serve({
port: process.env.PORT || 8080,
development: false,
// Share the same port across multiple processes
// This is the important part!
reusePort: true,
async fetch(request) {
return new Response("Hello from Bun #" + id + "!\n");
}
});
Linux only — Windows and macOS ignore the reusePort option. This is an operating system limitation with SO_REUSEPORT, unfortunately.
After saving the file, start your servers on the same port.
Under the hood, this uses the Linux SO_REUSEPORT and SO_REUSEADDR socket options to ensure fair load balancing across multiple processes. Learn more about SO_REUSEPORT and SO_REUSEADDR
import { spawn } from "bun";
const cpus = navigator.hardwareConcurrency; // Number of CPU cores
const buns = new Array(cpus);
for (let i = 0; i < cpus; i++) {
buns[i] = spawn({
cmd: ["bun", "./server.ts"],
stdout: "inherit",
stderr: "inherit",
stdin: "inherit",
});
}
function kill() {
for (const bun of buns) {
bun.kill();
}
}
process.on("SIGINT", kill);
process.on("exit", kill);
At the time of writing, Bun hasn't implemented the node:cluster module yet, but this is a faster, simple, and limited alternative. We will also implement node:cluster in the future.