locally

O

Anyone able to create an app with the new Remix worker template? I tried following https://developers.cloudflare.com/workers/frameworks/framework-guides/remix/ but the app it made seems to still be using pages. eg, the deploy command it gives is

wrangler pages deploy

wrangler pages deploy

wrangler pages deploy

wrangler pages deploy and I don't see an assets binding anywhere and it's still using @remix-run/cloudflare-pages...

Cloudflare Docs

Remix | Cloudflare Workers docs

Create a Remix application and deploy it to Cloudflare Workers with Workers Assets.

J

Jonathan Silva•10/3/24, 12:51 AM

I need myblogexample.pages.dev to be accessed by myblogexample.org but when I add the domain in the Pages settings, the message below appears

"After the update, visitors will no longer be able to access your site when visiting https://myblogexample.org."

I didn't understand

I want it to be accessed by myblogexample.org without displaying myblogexample.pages.dev

JJonathan Silva I need myblogexample.pages.dev to be accessed by myblogexample.org but when I ad...

H

Hello, I’m Allie!•10/3/24, 12:54 AM

It sounds like you might already have a DNS Record for that domain? I think it is warning you that it would remove that DNS record

HHello, I’m Allie!It sounds like you might already have a DNS Record for that domain? I think it i...

J

Jonathan Silva•10/3/24, 12:58 AM

Yes

4 types:
R2 ( bucket )
CNAME ( www to myblogexample.org )
Worker ( subdomain.myblogexample.org )
Worker ( subdomainb.myblogexample.org )

FFlikTeoh Because you’re using wrangler PAGES deploy

O

outsideurimagination•10/3/24, 1:34 AM

Yes, I'm saying the command in that doc

npm create cloudflare@latest my-remix-app -- --framework=remix --experimental

npm create cloudflare@latest my-remix-app -- --framework=remix --experimental

npm create cloudflare@latest my-remix-app -- --framework=remix --experimental

npm create cloudflare@latest my-remix-app -- --framework=remix --experimental generates that but it shouldn't.

SSia I just want to say this was a lovely read: https://blog.cloudflare.com/more-npm-...

R

Ricky U•10/3/24, 5:34 AM

seems like still no mongo support?

VVoyage Bagage I have one bot made the same way in JS that responding fine with telegraf api, b...

V

Voyage Bagage•10/3/24, 6:39 AM

V

Voyage Bagage•10/3/24, 6:39 AM

Anyone has ever had issues building telegram bot where the worker is not responding , or is but a while later ?
my previous bot was working fine but don't know what is wrong with that one ...

F

Flare•10/3/24, 7:02 AM

Please do not post your question in multiple channels/post it multiple times per the rules at ⁠#

welcome-and-rules. It creates confusion for people trying to help you and doesn't get your issue or question solved any faster.

A

Adi•10/3/24, 7:33 AM

Hey all, I just went through the process of migrating my Remix site to the new Workers static assets setup.
Things went pretty smoothly, however the one thing I couldn't find a solution for is how to set the

headers

headers

headers

headers for various asset routes.
In pages, we used to be able to have a

_headers

_headers

_headers

_headers file in the public folder - https://developers.cloudflare.com/pages/configuration/headers/
But that doesn't seem to work anymore.
What is the recommended way to do this now?

Cloudflare Docs

Headers | Cloudflare Pages docs

To attach headers to Cloudflare Pages responses, create a _headers plain text file in the output folder of your project. It is usually the folder that contains the deploy-ready HTML files and assets generated by the build, such as favicons. Changes to headers will be updated to your website at build time. Make sure you commit and push the file t...

AAdi Hey all, I just went through the process of migrating my Remix site to the new W...

H

Hello, I’m Allie!•10/3/24, 7:37 AM

Afaik currently you would have to apply headers via a rule(or via the Worker itself)

HHello, I’m Allie!Afaik currently you would have to apply headers via a rule(or via the Worker its...

A

Adi•10/3/24, 7:41 AM

Hmm I see. Ok, thanks for clarifying. I'll look into how to set the headers on the worker directly.

NNoireFatale locally

A

Ashley•10/3/24, 8:47 AM

Ah, are you looking for the logs on the dashboard in the browser? As logs for locally-running apps will be displayed in your terminal (where you wrangler dev / npm run dev)

Ooutsideurimagination Yes, I'm saying the command in that doc `npm create cloudflare@latest my-remix-a...

P

petebd•10/3/24, 9:41 AM

Hey @outsideurimagination - that create-cloudflare usage does actually create a Workers app, it's just that the npm "start" script has not been updated to Workers (it incorrectly still mentions Pages).
But you will not that the "deploy" and "preview" scripts do actually result in workers:

    "build": "remix vite:build && wrangler pages functions build --outdir build/worker",
    "deploy": "npm run build && wrangler deploy",
    "preview": "npm run build && wrangler dev",

    "build": "remix vite:build && wrangler pages functions build --outdir build/worker",
    "deploy": "npm run build && wrangler deploy",
    "preview": "npm run build && wrangler dev",

    "build": "remix vite:build && wrangler pages functions build --outdir build/worker",
    "deploy": "npm run build && wrangler deploy",
    "preview": "npm run build && wrangler dev",

    "build": "remix vite:build && wrangler pages functions build --outdir build/worker",
    "deploy": "npm run build && wrangler deploy",
    "preview": "npm run build && wrangler dev",

It is slightly confusing because it is indeed still using the Cloudflare Pages Remix adaptor (for the moment) which generates a Pages Function output, which then needs to be converted to a Worker via the call to

wrangler pages functions build

wrangler pages functions build

wrangler pages functions build

wrangler pages functions build.

I hope to have a better template out soon

Ppetebd Hey @outsideurimagination - that create-cloudflare usage does actually create a ...

O

outsideurimagination•10/3/24, 12:10 PM

ok thanks. I know they’re also updating the template in the remix repo. might check that out if you haven’t to not duplicate effort https://github.com/remix-run/remix/pull/10034

GitHub

templates(cloudflare-workers): use static assets by edmundhung · Pu...

This PR updates the cloudflare-workers template to use the Static Assets feature announced today.

Note: Static assets is currently in open beta

You can learn more about Static Assets in the docum...

F

Foreignity•10/3/24, 1:38 PM

I'm trying to access Origin from a function, but it doesn't return anything. Is it possible to access the Origin of a request, if so, what am I doing wrong?

export async function onRequest(context) {
const origin = context.request.headers.get('Origin') || 'Not available';
...
}

B

berkinovish•10/3/24, 5:20 PM

Any issues with workers?

ctx

ctx

seems to be empty for us sometimes, since a couple of hours

Bberkinovish Any issues with workers? `ctx` seems to be empty for us sometimes, since a coupl...

D

DaniFoldi•10/3/24, 5:37 PM

I created a test worker, seems fine from a few colos that I tested

DDaniFoldi I created a test worker, seems fine from a few colos that I tested

B

berkinovish•10/3/24, 6:19 PM

Seems that something changed under the hood in workers - we did a check to see if ctx was a valid object somewhere in our code, and this started failing since a couple of hours. I guess a change in the latest workerd that Cloudflare started rolling out globally

D

DaniFoldi•10/3/24, 6:19 PM

can you share the code for this check?

D

DaniFoldi•10/3/24, 6:20 PM

I log the typeof ctx and the constructor.toString, and it looks correct, and identical across regions

DDaniFoldi can you share the code for this check?

B

berkinovish•10/3/24, 6:21 PM

We used a util that (among other things) checked for this under the hood - this is the part that failed

    for (const _ in val) {
        return true;
    }

    for (const _ in val) {
        return true;
    }

    for (const _ in val) {
        return true;
    }

    for (const _ in val) {
        return true;
    }

U

un1que1•10/3/24, 6:21 PM

Hello everyone, I need help setting up multiple Shopify stores on a single .com domain using Cloudflare. The idea is to use Workers or Transform Rules to automatically switch between different Shopify subdomains at set intervals (e.g., 12 hours per store) for load balancing. The main domain should always mask the Shopify subdomains, with all routing done through Cloudflare, without any changes on Shopify's end. Each store will have the same content. How can I achieve this and is it even possible?

Bberkinovish We used a util that (among other things) checked for this under the hood - this ...

D

DaniFoldi•10/3/24, 6:32 PM

could you share a little more so I can repro and escalate if it is actually a runtime change?

L

louiek22•10/3/24, 8:43 PM

Has anyone else seen a significant jump in bundle size when using a compatibility date >=

2024-09-23

2024-09-23

2024-09-23

2024-09-23 with the

nodejs_compat

nodejs_compat

nodejs_compat

nodejs_compat flag? We saw about a 2x increase in bundle size on our workers.

K

kian•10/3/24, 8:46 PM

That's intentional -

nodejs_compat

nodejs_compat

nodejs_compat

nodejs_compat implies

nodejs_compat_v2

nodejs_compat_v2

nodejs_compat_v2

nodejs_compat_v2 after that

compatibility_date

compatibility_date

compatibility_date

compatibility_date & Wrangler adds polyfills with unenv

DDaniFoldi could you share a little more so I can repro and escalate if it is actually a ru...

B

berkinovish•10/3/24, 8:55 PM

Yes, this crashes with

"wrangler": "3.78.10"

"wrangler": "3.78.10"

, but works fine with wrangler on

3.78.9

3.78.9

:

function isNotEmptyObject<T extends Record<string, any>>(val: T | unknown): val is T {
    if (Object.prototype.toString.call(val) !== '[object Object]') return false;
    for (const _ in val as Record<string, any>) {
        return true;
    }
    return false;
}

export default {
    async fetch(request, env, ctx): Promise<Response> {
        if (!isNotEmptyObject(ctx)) throw new Error('ctx');
        return new Response('Hello World!');
    },
} satisfies ExportedHandler<Env>;

function isNotEmptyObject<T extends Record<string, any>>(val: T | unknown): val is T {
    if (Object.prototype.toString.call(val) !== '[object Object]') return false;
    for (const _ in val as Record<string, any>) {
        return true;
    }
    return false;
}

export default {
    async fetch(request, env, ctx): Promise<Response> {
        if (!isNotEmptyObject(ctx)) throw new Error('ctx');
        return new Response('Hello World!');
    },
} satisfies ExportedHandler<Env>;

function isNotEmptyObject<T extends Record<string, any>>(val: T | unknown): val is T {
    if (Object.prototype.toString.call(val) !== '[object Object]') return false;
    for (const _ in val as Record<string, any>) {
        return true;
    }
    return false;
}

export default {
    async fetch(request, env, ctx): Promise<Response> {
        if (!isNotEmptyObject(ctx)) throw new Error('ctx');
        return new Response('Hello World!');
    },
} satisfies ExportedHandler<Env>;

function isNotEmptyObject<T extends Record<string, any>>(val: T | unknown): val is T {
    if (Object.prototype.toString.call(val) !== '[object Object]') return false;
    for (const _ in val as Record<string, any>) {
        return true;
    }
    return false;
}

export default {
    async fetch(request, env, ctx): Promise<Response> {
        if (!isNotEmptyObject(ctx)) throw new Error('ctx');
        return new Response('Hello World!');
    },
} satisfies ExportedHandler<Env>;

We think this might be the commit that caused the behaviour change: https://github.com/cloudflare/workerd/pull/2801/files

I assume impact will be low though, pretty niche to do this check on

ctx

ctx

Llouiek22 Has anyone else seen a significant jump in bundle size when using a compatibilit...

D

DaniFoldi•10/3/24, 8:56 PM

that's because workers with nodejs_compat and a compat date that is newer than your date have some js polyfills added, in addition to the runtime compat mode that it enables

Bberkinovish Yes, this crashes with `"wrangler": "3.78.10"`, but works fine with wrangler on ...

D

DaniFoldi•10/3/24, 8:56 PM

thanks, I'll forward this info

Kkian That's intentional - `nodejs_compat` implies `nodejs_compat_v2` after that `comp...

L

louiek22•10/3/24, 8:58 PM

Gotcha, I thought unenv was supposed to be smart about adding polyfills. Thanks for the info!

DDaniFoldi thanks, I'll forward this info

B

berkinovish•10/3/24, 8:58 PM

Even though we fixed our wrangler at 3.78.6, we were still impacted - it looked like there was a gradual rollout of a newer version throughout the day

Llouiek22 Gotcha, I thought unenv was supposed to be smart about adding polyfills. Thanks ...

D

DaniFoldi•10/3/24, 8:58 PM

it should only add what you might use

Bberkinovish Even though we fixed our wrangler at 3.78.6, we were still impacted - it looked ...

D

DaniFoldi•10/3/24, 8:59 PM

correct, that's usually how runtime releases are done, and your version of wrangler only affects your local runtime version

DDaniFoldi correct, that's usually how runtime releases are done, and your version of wrang...

B

berkinovish•10/3/24, 8:59 PM

makes sense

DDaniFoldi it should only add what you _might_ use

L

louiek22•10/3/24, 8:59 PM

I'm not using anything in the worker itself, just have it for vitest.

L

louiek22•10/3/24, 9:01 PM

I'm guessing that extra size is from boilerplate from unenv, or maybe it's coming from my tests?

R

rufi•10/3/24, 11:59 PM

how many requests can a single worker handle? in AWS, GCloud and al., we see the concept of "quotas" and number of "container instances" spawned over time. We don't have such information on Cloudflare dashboard. Is it using a single worker for each request; or rather a single worker can handle millions or more requests by itself?

W

Walshy•10/4/24, 12:14 AM

There is no limit, I've done over 100k rps myself from my own laptop and of course we do much more as a whole platform

Each Worker can handle tons and there could many many thousands of worker instances.

Workers are light and so can be created everywhere and handle a lot

A

Akash•10/4/24, 1:14 AM

does snippets support kv? If not is there any plan to add this support?

A

Ashish Rawat•10/4/24, 5:28 AM

I wish workers has support for firebase sdk and mongodb. That's the only thing stopping me to deploy to workers. Currently have to use google cloud run for docker instance.

AAshish Rawat I wish workers has support for firebase sdk and mongodb. That's the only thing s...

H

hfcRed•10/4/24, 5:50 AM

Why not D1 or KV

AAkash does snippets support kv? If not is there any plan to add this support?

W

Walshy•10/4/24, 8:47 AM

Snippets have no bindings access. They're meant for lightweight use cases

W

w3bcode•10/4/24, 9:03 AM

Anyone getting error with wrangler deploy command, I am latest version 3.79.0

workers.api.error.unknown [code: 10013]

Ww3bcode Anyone getting error with wrangler deploy command, I am latest version 3.79.0 > ...

W

Walshy•10/4/24, 9:34 AM

What's your account ID?

W

w3bcode•10/4/24, 1:31 PM

9c508e2e1d1c6d075d774819cf38176c

I was using it with Cloudflare account token template settings by exposing it to environment but getting api errors, so I logged in with wrangler login and deployed. It worked that way

F

Focux•10/4/24, 1:35 PM

is there any template for running remix spa mode in workers?

HHello, I’m Allie!Afaik currently you would have to apply headers via a rule(or via the Worker its...

F

Focux•10/4/24, 1:43 PM

If the _headers file doesn't work anymore, would be good to remove it from the template, otherwise, it's a bit confusing.

Similar Threads

Similar Threads

Similar Threads