Can you use durable objects on workers for platforms? Can user workers and dynamic dispatch workers
Can you use durable objects on workers for platforms? Can user workers and dynamic dispatch workers both access them?
SWas wondering that as well. The docs are stating any bindings would work, but not sure if just from the main account or from user namespaces as well, I.e. deployed / created on behalf of the user
Sok, I think that's probably the answer https://developers.cloudflare.com/kv/examples/workers-kv-to-serve-assets/
Cloudflare Docs
Example of how to use Workers KV to store static assets
PYes you can. User workers can host DOs, and access their own DOs or DOs from account-level workers. Accessing other user workers DOs is either unsupported or not documented.
To solve that limitation in our setup we use a 2nd user worker that hosts all the DOs, and do some magic with service bindings so the original user worker, our dispatcher and other user scripts can access them.
To solve that limitation in our setup we use a 2nd user worker that hosts all the DOs, and do some magic with service bindings so the original user worker, our dispatcher and other user scripts can access them.
1const client = new Cloudflare({
apiToken: process.env["CLOUDFLARE_API_TOKEN"],
});
const zones = await client.zones.list();
console.log(zones);
apiToken: process.env["CLOUDFLARE_API_TOKEN"],
});
const zones = await client.zones.list();
console.log(zones);
Pdash is always kinda sluggish
Pbut I haven't noticed any issues specifically with CF Workers
Pare you noticing it specifically with Workers for Platforms?
B#workers-discussions is probably a better bet
RFor me radar.cloudflare.com is very often slow, but that might be also some peering things with the Deutsche Telekom (german ISP)
RAre Workers Builds coming to Workers for Platforms? For my use case this would be helpful
TWhere is this documented?
PIt's the exact same as regular workers
JHi guys,
I have proxied the workers custom domain with the https://flexiff.saurabhbhatnagar.in/8zypxuri domain, if i published new content in the page, cache was not cleared suddenly, it is taking 4 hours default cloudflare cache.
https://flexiff.saurabhbhatnagar.in this domain
I have proxied the workers custom domain with the https://flexiff.saurabhbhatnagar.in/8zypxuri domain, if i published new content in the page, cache was not cleared suddenly, it is taking 4 hours default cloudflare cache.
https://flexiff.saurabhbhatnagar.in this domain
test
FlexiFunnels
Jmay i know how to clear the cache in the worker?
Amajor blocker on static assets for now, eagerly waiting 
any workaround? (trying to take advantage of frameworks on workers :D)
any workaround? (trying to take advantage of frameworks on workers :D)
PYou could develop your own assets fetcher using KV or R2, but it's not the same unfortunately
SIs there an updated doc for the limitations on WFP? What features are limited or not available in user workers (Durable Objects, Workflow, queues, etc)?
Cgood morinng
Cmorning !!
DHello, we're building a functions runtime on top of workers-for-platforms. We've built an API that allows users to deploy a function script, and then execute it either by an HTTP call, or based on a trigger like a cron job or a database mutation.
Currently we're struggling with testing this system. As far as I know there's no way to deploy a worker to a dispatch-namespace locally.
Any suggestions from the Cloudflare team on how to test dispatch-namespaces locally would be appreciated.
Currently we're struggling with testing this system. As far as I know there's no way to deploy a worker to a dispatch-namespace locally.
Any suggestions from the Cloudflare team on how to test dispatch-namespaces locally would be appreciated.
D@tanushree │ Workers PM We'd love to connect with someone from the workers-for-platforms team to share feedback and ask some questions about the product.
Pquestion, here, is there a way to do a spend cap, example like number of requests received?
https://developers.cloudflare.com/cloudflare-for-platforms/workers-for-platforms/configuration/custom-limits/
https://developers.cloudflare.com/cloudflare-for-platforms/workers-for-platforms/configuration/custom-limits/
Cloudflare Docs
Custom limits allow you to programmatically enforce limits on your customers' Workers' resource usage. You can set limits for the maximum CPU time and number of subrequests per invocation. If a user Worker hits either of these limits, the user Worker will immediately throw an exception.
SHi team , just hopped around cloudfare workers wondering , if there is a way to deploy my own models into workers ai that huggingface do not support direct deployment to workers ai currently . And if I wanted to re-train that model privately on my data what would the process look like
PHi everyone. I started to use the workers-for-plataforms today and I'm having some trouble regarding env variables.
My dispatcher does see all my enviroment variables, but my worker that is in the same namespace does not recieve any variables in the "env" parameter of the fetch function. Does any one know how to fix that?
My dispatcher does see all my enviroment variables, but my worker that is in the same namespace does not recieve any variables in the "env" parameter of the fetch function. Does any one know how to fix that?
SI know it's not available right now, but does anyone know whether Workers Static Assets is on the roadmap for WFP?
WYes it is 
SExcellent! Good to know, thanks
MHello guys, I am working on a project with dynamically uploaded user workers, and I am trying to write unit test that works locally with
I am using wrangler version
Test scenario is inspired by https://developers.cloudflare.com/cloudflare-for-platforms/workers-for-platforms/get-started/developing-with-wrangler/ and looks like that:
0) Start
1) Start
2) "Dispatch worker" makes
Example of user worker's config:
3) After mocked upload, "dispatch worker" is doing
On this step I am getting error response
So I do not understand why my mocked
DISPATCH_NAMESPACE binding. I am using wrangler version
0.0.0-2ca4eaef5 with tag dispatch-namespaces-dev as recommended in workers for platforms documentation.Test scenario is inspired by https://developers.cloudflare.com/cloudflare-for-platforms/workers-for-platforms/get-started/developing-with-wrangler/ and looks like that:
0) Start
unstable_dev "outbound worker".1) Start
unstable_dev "dispatch worker" that has DISPATCH_NAMESPACE binding with outbound = { service = "outbound-worker" } 2) "Dispatch worker" makes
PUT request <ACCOUNT_ID>/workers/dispatch/namespaces/<NAMESPACE>/scripts/<SCRIPT_NAME> to my local nodejs sever that spawns child process with wrangler devExample of user worker's config:
3) After mocked upload, "dispatch worker" is doing
const worker = DISPATCH_NAMESPACE.get(<SCRIPT_NAME>); worker.fetch(...) to dispatch user worker.On this step I am getting error response
[wrangler] Couldn\'t find 'wrangler dev' session for "0193d5cc-506a-77db-a12a-c81bd1d6b6dc-0-0-0" in namespace "dev-functions" to proxy to. And no request is being intercepted by my outbound worker.So I do not understand why my mocked
user workers is not being seen by dispatch worker, maybe I should not be using unstable_dev for starting dispatch worker. Would appreciate any help, thanks in advance.Cloudflare Docs
To test your Dispatch Worker, user Worker and Outbound Worker before deploying to production, you can use Wrangler for development and testing.
Mcc @Walshy Nice to meet you, just trying to get someone's attention here
MHi everyone! 
I'm testing out Workers for my platform and currently building a dispatch worker with dynamic namespace bindings that are added or removed via API. However, I have a question about managing namespace bindings during deployment.
In my CI/CD setup for the dispatch worker, I have a wrangler.toml file, but I don’t want it to override the dynamic namespace bindings created through the API.
Is there a way to handle this using Wrangler? Or should I rely solely on the API for this scenario? Any advice would be greatly appreciated!
I'm testing out Workers for my platform and currently building a dispatch worker with dynamic namespace bindings that are added or removed via API. However, I have a question about managing namespace bindings during deployment.
In my CI/CD setup for the dispatch worker, I have a wrangler.toml file, but I don’t want it to override the dynamic namespace bindings created through the API.
Is there a way to handle this using Wrangler? Or should I rely solely on the API for this scenario? Any advice would be greatly appreciated!
MNvm, I'm just misusing dispatch namespaces
MWhy it is written in documentation:
We discourage creating a new namespace for each customer?WBecause 1 namespace can handle infinite scripts
Wbut you need 1 binding per namespace
WIt provides no benefit security or otherwise but massively impacts dx on your side
MCause they need to be set in wrangler.toml?
WYes, you need a binding per namespace
MCan't I use API to abstract this constraint, just pushing the script instead of the whole bundle?
WNope, I'm not sure how you'd go about doing that
MSo updating my code will need me to update dynamically my bindings after?
WLet's step back - why do you want a namespace per customer?
MIt's easier to manage customers like this instead based on tags, it'll be quicker to develop our backend
DWhat's the latency look like with WFP compared to Service Bindings?
DI know the Service Bindings docs say no added latency... curious how WFP differs
