workers-sdk: 🐛 BUG: fetching bound registered workers in local mode fails with timeout

What version of `Wrangler` are you using?

2.1.6

What operating system are you using?

M1 Pro macOS 12.6

Describe the Bug

When running wrangler pages dev, after some delay (presumably the timeout mentioned in the error) this error is being output several times per second until the server is stopped.

▲ [WARNING] Failed to get worker definitions TypeError: fetch failed

      at Object.processResponse
  (/Users/andre/code/wrangler-bug/node_modules/wrangler/wrangler-dist/cli.js:8846:27)
      at /Users/andre/code/wrangler-bug/node_modules/wrangler/wrangler-dist/cli.js:9178:42
      at node:internal/process/task_queues:140:7
      at AsyncResource.runInAsyncScope (node:async_hooks:203:9)
      at AsyncResource.runMicrotask (node:internal/process/task_queues:137:8)
      at process.processTicksAndRejections (node:internal/process/task_queues:95:5) {
    cause: ConnectTimeoutError: Connect Timeout Error
        at onConnectTimeout
  (/Users/andre/code/wrangler-bug/node_modules/wrangler/wrangler-dist/cli.js:2870:29)
        at /Users/andre/code/wrangler-bug/node_modules/wrangler/wrangler-dist/cli.js:2828:51
        at Immediate._onImmediate
  (/Users/andre/code/wrangler-bug/node_modules/wrangler/wrangler-dist/cli.js:2859:13)
        at process.processImmediate (node:internal/timers:471:21) {
      code: 'UND_ERR_CONNECT_TIMEOUT'
    }
  }

steps to reproduce:

npx create-remix@latest
- name
- Just The Basics
- Cloudflare Pages
- Typescript
- Y
npm run dev
wait 10-15 seconds.

Note: I do not have wrangler installed globally, and have tried running this same command after running npx wrangler login and the error is still output.

About this issue

Original URL
State: closed
Created 2 years ago
Reactions: 3
Comments: 17

Most upvoted comments

I fixed this on my machine by finding the port listening on 6284 (the dev registry server) and force quitting this. I’m on macos and performed the following steps:

sudo lsof -i -P | grep LISTEN | grep :6284

This outputted the process name (“node”) and the PID. I then found this process in Activity Monitor and force quit it.

Restarting your machine would probably do a similar job.

tom-sherman on Nov 26, 2022

This seemed to work for me: kill -9 $(lsof -ti:6284)

It is my understanding that this will find and kill the node process of the Worker Registry.

rosswaycaster on Jul 12, 2023

Then I can kill the worker

You need to kill more than the worker. There’s a seperate background process (dev registry server) that you need to kill - it should be on port 6284.

tom-sherman on Feb 9, 2023

I’m encountering this error as well on 2.3.2

cmoog on Nov 16, 2022