Hey folks, thinking about using CF Workers For Platform to build an Integrations Platform, and I was

Ppravin Hey folks, thinking about using CF Workers For Platform to build an Integrations...

R

rdutton•1/7/25, 2:31 AM

One of the challenges with webhooks is sending a response within the required timeframe. e.g. Shopify and Xero will only wait 5 seconds. Assuming your application processing could take longer that 5 seconds, two options are: 1) Respond early and continue execution using 'waitUntil'. see https://developers.cloudflare.com/workers/runtime-apis/context/#waituntil or 2) Grab the request data (optionally store it) and send it into a queue for later processing by another worker, then respond. Personally I use the second pattern as it provides more visibility and the opportunity to replay the webhook event if needed.

Rrdutton One of the challenges with webhooks is sending a response within the required ti...

P

pravinOP•1/7/25, 5:31 AM

Yep, makes sense, thanks @rdutton

I didn't see Queues in the WfP documentation. I suppose everything that works with Workers will also work in a WfP setting?

Ppravin Yep, makes sense, thanks @rdutton I didn't see Queues in the WfP documentation...

R

rdutton•1/7/25, 5:37 AM

I haven't used WFP before so can't say for sure. Someone else might be able to confirm.

C

christianklotz•1/8/25, 2:02 PM

I followed the guide which uses

wranger@dispatch-namespaces-dev

wranger@dispatch-namespaces-dev

wranger@dispatch-namespaces-dev

wranger@dispatch-namespaces-dev but local dev environment fails with

[wrangler] Couldn't find `wrangler dev` session for "customer-worker-1" in namespace "workers-for-platforms-playground" to proxy to

[wrangler] Couldn't find `wrangler dev` session for "customer-worker-1" in namespace "workers-for-platforms-playground" to proxy to

[wrangler] Couldn't find `wrangler dev` session for "customer-worker-1" in namespace "workers-for-platforms-playground" to proxy to

[wrangler] Couldn't find `wrangler dev` session for "customer-worker-1" in namespace "workers-for-platforms-playground" to proxy to

Can someone confirm that this is actually supported in this wrangler dev branch release? Or is uploading scripts the only way to debug WFP apps?

S

sherlock•1/9/25, 6:22 AM

I added a custom domain and when I access it through the custom domain, I get this error

error code: 1003

S

sherlock•1/9/25, 6:24 AM

help me

A

arun2728•1/9/25, 7:59 AM

I am getting this error when creating a new cloudflare worker - workers.api.error.exceeded_allowed_number_of_scripts. When I checked the cloudflare docs it suggested to upgrade to Cloudflare Workers for Platforms. And I upgraded the plan. But I am still getting this error. Can anyone help me with it?

W

Walshy•1/9/25, 1:49 PM

Workers for Platforms is indeed the solution, you'll want to switch to uploading the scripts as Workers for Platforms scripts rather than regular Workers

W

Walshy•1/9/25, 1:50 PM

https://developers.cloudflare.com/api/resources/workers_for_platforms/subresources/dispatch/subresources/namespaces/subresources/scripts/methods/update/

W

Walshy•1/9/25, 5:07 PM

For #1 you should follow the same as #2, add them to ssl for saas and add a CNAME to your main zone

W

Walshy•1/9/25, 5:08 PM

And yeah, for you it seems you can just keep using 1 worker, let it get invoked by ssl for saas and it'll be handled for all customers with ease

W

Walshy•1/9/25, 5:19 PM

I don't think there is a limit on hostnames?

W

Walshy•1/9/25, 5:19 PM

Workers for Platforms wouldn't grant you anything extra but would cost more

MMatt yeah there is: https://developers.cloudflare.com/cloudflare-for-platforms/cloudf...

W

Walshy•1/9/25, 5:20 PM

huh til

W

Walshy•1/9/25, 5:20 PM

ssl for saasing this fully is the only scalable solution unfortunately.

W

Walshy•1/9/25, 5:21 PM

We have a limit on the amount of zones a Worker can be linked to, I think it's 100 so you'd hit that much quicker than the saas limit

W

Walshy•1/9/25, 5:22 PM

You only need 1 route -

*/*

*/*

W

Walshy•1/9/25, 5:22 PM

That will accept all hostnames and all paths

J

juanferreras•1/9/25, 8:42 PM

has anyone tried to use some of https://workers.cloudflare.com/playground for users to write WfP User Workers? I see most of the code seems open source, but wondering if I'll go down an unescapable rabbit hole

Jjuanferreras has anyone tried to use some of https://workers.cloudflare.com/playground for us...

J

juanferreras•1/15/25, 10:26 PM

ended up using https://github.com/val-town/codemirror-ts with pretty good results (you do need to upload the built output to wrangler) for a single-file user worker.

it's pretty good but still kind of missing a way to let users add a dependency – it'd be pretty cool to defer this to the API or to

wrangler

wrangler

wrangler

wrangler somehow to build/bundle correctly

Jjuanferreras ended up using https://github.com/val-town/codemirror-ts with pretty good result...

H

Hello, I’m Allie!•1/16/25, 4:11 AM

You could probably go the Stackblitz route? Only thing there is you can’t run dev there

T

tornado•1/23/25, 5:34 PM

Can I use workers-for-platforms to dynamically create vectorize database per customer?

A

arun2728•1/27/25, 9:53 AM

Hi there, I am using cloudflare workers for platform. I want to create a Create a dynamic dispatch Worker that can route the request to the correct worker by extracting the details from the hostname.

Eg: My dispatch workers should be able to accept all the requests for the domain:

*.route.example.com

*.route.example.com

*.route.example.com

*.route.example.com And route it to the child worker using the subdomain.

testing-1.route.example.com -> dispatch worker 1
testing-2.route.example.com -> dispatch worker 2

This is my sample routing script from the docs: https://developers.cloudflare.com/cloudflare-for-platforms/workers-for-platforms/get-started/dynamic-dispatch/

But docs doesn't specify how to setup domains to make it work. I am unable to figure out how to set the route or domain to my dispatcher worker so that I can accept requests. How can I do this? Can anyone help me with this?

export default {
  async fetch(request, env) {
    try {
      // parse the URL, read the subdomain
      let workerName = new URL(request.url).host.split('.')[0];
      let userWorker = env.dispatcher.get(workerName);
      return await userWorker.fetch(request);
    } catch (e) {
      if (e.message.startsWith('Worker not found')) {
        // we tried to get a worker that doesn't exist in our dispatch namespace
        return new Response('', { status: 404 });
      }

      // this could be any other exception from `fetch()` *or* an exception
      // thrown by the called worker (e.g. if the dispatched worker has
      // `throw MyException()`, you could check for that here).
      return new Response(e.message, { status: 500 });
    }
  },
};

export default {
  async fetch(request, env) {
    try {
      // parse the URL, read the subdomain
      let workerName = new URL(request.url).host.split('.')[0];
      let userWorker = env.dispatcher.get(workerName);
      return await userWorker.fetch(request);
    } catch (e) {
      if (e.message.startsWith('Worker not found')) {
        // we tried to get a worker that doesn't exist in our dispatch namespace
        return new Response('', { status: 404 });
      }

      // this could be any other exception from `fetch()` *or* an exception
      // thrown by the called worker (e.g. if the dispatched worker has
      // `throw MyException()`, you could check for that here).
      return new Response(e.message, { status: 500 });
    }
  },
};

export default {
  async fetch(request, env) {
    try {
      // parse the URL, read the subdomain
      let workerName = new URL(request.url).host.split('.')[0];
      let userWorker = env.dispatcher.get(workerName);
      return await userWorker.fetch(request);
    } catch (e) {
      if (e.message.startsWith('Worker not found')) {
        // we tried to get a worker that doesn't exist in our dispatch namespace
        return new Response('', { status: 404 });
      }

      // this could be any other exception from `fetch()` *or* an exception
      // thrown by the called worker (e.g. if the dispatched worker has
      // `throw MyException()`, you could check for that here).
      return new Response(e.message, { status: 500 });
    }
  },
};

export default {
  async fetch(request, env) {
    try {
      // parse the URL, read the subdomain
      let workerName = new URL(request.url).host.split('.')[0];
      let userWorker = env.dispatcher.get(workerName);
      return await userWorker.fetch(request);
    } catch (e) {
      if (e.message.startsWith('Worker not found')) {
        // we tried to get a worker that doesn't exist in our dispatch namespace
        return new Response('', { status: 404 });
      }

      // this could be any other exception from `fetch()` *or* an exception
      // thrown by the called worker (e.g. if the dispatched worker has
      // `throw MyException()`, you could check for that here).
      return new Response(e.message, { status: 500 });
    }
  },
};

Cloudflare Docs

Create a dynamic dispatch Worker · Cloudflare for Platforms docs

After you have created a dispatch namespace, you can fetch any user Workers in the namespace using a dynamic dispatch Worker. The dynamic dispatch Worker has a namespace binding.

DDORSHT Hey, beginner question that maybe someone here can help me out with. I'm going t...

A

arun2728•1/27/25, 10:38 AM

@DORSHT Were you able to implement this? I am facing the same issue. And I am not able to figure it out.

S

safah222•1/28/25, 3:31 PM

Hello , guys , how i can exclude specific path for a worker route , as i want it to run on wordpress website , how ever triggering it as wild cart , will let the worker run millions of request for wp-admin , and wp-content reuqests , so i want to exclude wp-content from the routes how i can do that
why no one anwser please help itsurgent

Y

yags•1/29/25, 5:29 PM

Hello, are there known issues with workers right now? I'm getting 503 on deployment

 [ERROR] Received a malformed response from the API

  upstream connect error or disconnect/reset before headers. reset reason: connection termination

 [ERROR] Received a malformed response from the API

  upstream connect error or disconnect/reset before headers. reset reason: connection termination

 [ERROR] Received a malformed response from the API

  upstream connect error or disconnect/reset before headers. reset reason: connection termination

 [ERROR] Received a malformed response from the API

  upstream connect error or disconnect/reset before headers. reset reason: connection termination

Y

yags•1/29/25, 5:31 PM

^ after many attempts, worked

M

Mercie•1/29/25, 5:31 PM

I'm also getting tons of 500s and 503s in the web ui for all kinds of things, it does look like an outage to me

M

Mercie•1/29/25, 5:32 PM

trying to massage my workers haha

W

Walshy•1/29/25, 5:38 PM

Should be good again - sorry about that

M

Mercie•1/29/25, 5:39 PM

Thanks!

M

Mercie•1/29/25, 5:39 PM

Looks good to me

N

neoX•1/30/25, 6:51 PM

I changed my hosting server from bluehost to cloudflare for ssl/tls and free CDN service. I did updated the dns records to cloudflare ones and I also stopped auto-ssl from bluehost but still I'm unable to use Cloudflare ssl. (Should i need to delete bluhost ssl certificates???)

Screenshot_2025-01-30_at_11.42.05_PM.png

R

rob•1/31/25, 3:02 PM

Can you bind a specific gateway to a worker on wfp!

R

rob•1/31/25, 3:02 PM

Or does binding it to AI give it access to all gateways?

A

Advany•2/1/25, 8:25 PM

We have customers developing a workflow as a worker using ai. We deploy that to wfp as a worker with endpoints like /step1 to /step10.

We where thinking of using a durable object to call the worker thats deployed. For example first step1, get result, then call step2, etc. Untill done.

Is there a better way? Steps can take while waiting for external apis like ai

E

Exception•2/4/25, 12:07 AM

What is the best way to go to update all workers' code in a particular namespace? Is there a command for that or will you need to do it in a loop?

EException What is the best way to go to update all workers' code in a particular namespace...

V

verdverm.com•2/4/25, 3:12 AM

there seem to be some breadcrumbs here, depending on the product you are using: https://developers.cloudflare.com/durable-objects/platform/known-issues/#code-updates

Cloudflare Docs

Known issues · Cloudflare Durable Objects docs

Durable Objects is generally available. However, there are some known issues.

Vverdverm.com there seem to be some breadcrumbs here, depending on the product you are using: ...

E

Exception•2/4/25, 4:09 AM

Thanks

D

Dalist•2/5/25, 12:29 PM

is there a way I can change the puppeteer browser's version? I'm getting 120 by default!

DDalist is there a way I can change the puppeteer browser's version? I'm getting 120 by ...

D

DaniFoldi•2/5/25, 6:17 PM

This isn't directly related to WfP, so I recommend asking in #workers-help or #workers-and-pages-discussions :meowwink:

A

apignotti•2/6/25, 9:23 AM

Morning all. I have a question about WfP and legal responsibility for outbound traffic from final users.

Let's assume a final user of my service (built on WfP) generate malicious/illegal traffic. It would be my responsibility to help law enforcement track the original user of the traffic. Does CF provide any builtin auditing/logging functionality to help in this or all WfP users must create a solutions from scratch?

Hey folks, thinking about using CF Workers For Platform to build an Integrations Platform, and I was

Similar Threads