LUsing envoy
LHave a wasm plugin that for eg. will return 401 if no api key is present
Lfor http1 this works, for grpc the LB returns 520
CWhat that sounds like to me is the gateway you are using is translating the GRPC -> to http1 error codes incorrectly
Cwhen the grpc request is authenticated does it work correctly?
LYep it does
Cyeah, so cloudflare part is working
LIf the code just responds with a 401, grpc-status header set, what else could be missing?
Cenvoy config could be wrong
PSo to use Pages advanced mode I have to use this https://developers.cloudflare.com/pages/platform/functions/advanced-mode/
I did follow the docs, but like im struggling with putting the _worker.js in the actual output dir??
Like as you can see in the first photo, running the build command generates a _worker.js file, but it is not the one that I wrote, how do I get MY worker to be in the build directory? Keep in mind I'm using svelte
I did follow the docs, but like im struggling with putting the _worker.js in the actual output dir??
Like as you can see in the first photo, running the build command generates a _worker.js file, but it is not the one that I wrote, how do I get MY worker to be in the build directory? Keep in mind I'm using svelte
PThis is my wrangler.toml file, it is worth noting that my worker is located in /worker/index.js
EIf you switch your Worker to the new Standard pricing (or even the old Unbound pricing), the subrequest limit will be 1000 instead of 50
ISvelteKit uses their own _worker.js, what is your purpose for using _worker.js? Also we should probably be talking in #functions
Palr we can move there
FI am using Remix deployed on Pages, but since it's SSR, I hear that this is actually running on workers
Fmy issue is that i want the cookies sent to remix to be sent to another server
Fis there any way to make this work?
Fyeah im kinda lost, i assumed using remix for the frontend would work and allow me to fetch from an api and send the whole response back, but the api requires the user to be logged in
CPut your account ID in browser-rendering-apiBrowser Rendering API Beta Access
0can anyone recommend a good tutorial that I can watch to get started with node.js serverless functions running on Cloudflare workers?
Lclass Zone {
constructor({ name, parent });
name;
get parent();
fork({ name });
run(callback);
wrap(callback);
static get current();
}
constructor({ name, parent });
name;
get parent();
fork({ name });
run(callback);
wrap(callback);
static get current();
}
N@Erisa | Support Engineer nice
NLooks like this document has been updated
HYou can’t run Node on Workers
HWorkers has some node-compatibility, but it generally isn’t just drag and drop
0I’m sorry I could be confused, I read in the docs that the node.js runtime is available
Maybe you can shed some light on this page here
https://developers.cloudflare.com/workers/runtime-apis/nodejs/#enable-nodejs-with-workers
Maybe you can shed some light on this page here
https://developers.cloudflare.com/workers/runtime-apis/nodejs/#enable-nodejs-with-workers
Most Workers import one or more packages of JavaScript or TypeScript code from npm as dependencies in package.json. Many of these packages rely on …
HThat page covers the Node compatibility layer. Some modules are supported, but Node as a whole isn’t
0Thanks for clarifying
I was told by multiple people that I could run langchain with Cloudflare workers & wrangler
Am I being led in the wrong direction?
The issue I’m having is that I want to run langchain in a react native mobile application but the Hermes engine does not use “readable streams” which is how langchain responds
I was told by multiple people that I could run langchain with Cloudflare workers & wrangler
Am I being led in the wrong direction?
The issue I’m having is that I want to run langchain in a react native mobile application but the Hermes engine does not use “readable streams” which is how langchain responds
HYou can run some models on Cloudflare with https://ai.cloudflare.com, but as to whether you can run Langchain itself, I'm not sure
MGuys, adding Durable Objects to my code causes it to 200ms-1s additional response times, Is that expected? Was trying to replace KV for authentication management
MEven when setting the locationHint
Swhat country
MBrazil, was trying to set locationHint to sam (South America)
CI'm pretty sure there is no Durable Object in South America 
This cool page shows where DOs are created: https://where.durableobjects.live
This cool page shows where DOs are created: https://where.durableobjects.live
Tracking where Durable Objects are created, wherever you are in the world.
MThat makes sense now! Thank you guys
MWill stick with KV for now, and durable for ws connections
H
GHi! The cloudflare API permit to create a Pages deployment from a repository, but I don't see any way from the API to do the same from a « Direct Upload » (.zip file) https://developers.cloudflare.com/api/operations/pages-deployment-create-deployment
Is it planned?
Is it planned?
Interact with Cloudflare's products and services via the Cloudflare API
CCan I fetch another worker in the same zone if my worker runs on a custom domain?
CLast I heard, you should be able to, the docs say so, but you can't unless you remove the miminimal_subrequests flag, ex: https://github.com/cloudflare/workerd/issues/787#issuecomment-1597396851
Service Bindings may be a better fit unless you really need the same-zone fetch: https://developers.cloudflare.com/workers/configuration/bindings/about-service-bindings/
They don't go over the internet, same isolate, so generally faster, and you can pass special properties over the request.cf property
Service Bindings may be a better fit unless you really need the same-zone fetch: https://developers.cloudflare.com/workers/configuration/bindings/about-service-bindings/
They don't go over the internet, same isolate, so generally faster, and you can pass special properties over the request.cf property
Service bindings are an API that facilitate Worker-to-Worker communication via explicit bindings defined in your configuration.
GitHub
Which Cloudflare product(s) does this pertain to? Workers/Other What version of Wrangler are you using? 2.9.0 What operating system are you using? Mac Describe the Bug Same zone worker <-> wo...
CThanks, that is what I am looking for. I encountered the same 522 error issue.
BI have 400 users details like email, name, contact etc inside an array. I want to insert/update all these 400 users profiles information inside 3rd party service called klaviyo for email marketing. In adding 1 user information into this 3rd party API, it takes 2 fetch requests (max 2 for update). So for 400 users its taking 800 fetch requests at a time for inserting. I am getting error inside logs that "subrequest limit exceeds" means so many subrequest a time...
What is solution of this, can anyone please help?
NOTE: I am using premium version not free
I have also attached a screenshot of my code with result array of 400 users and then adding them one by one into klaviyo
What is solution of this, can anyone please help?
NOTE: I am using premium version not free
I have also attached a screenshot of my code with result array of 400 users and then adding them one by one into klaviyo
AEither limit the number of customers you send per scheduled run (e.g. in your DB, have a column that is "sent" or whatever) and then retrieve X number per scheduler run and only pull ones where they aren't sent yet to the API. Or use https://developers.cloudflare.com/queues/, and push each customer as a message into a queue, and have a separate worker process the queue
AWell, those are two options anyway
AAI also don't know what it is for the new pricing model, as they only talk about bound vs unbound there