Anyone know why I'm getting 512 "Could not send your email." with the Mailchannels plugin? I believe

Anyone know why I'm getting 512 "Could not send your email." with the Mailchannels plugin? I believe I have my DNS record setup properly for the lockdown thing, it points to the subdomain in the "Pages & Workers Overview" on the right

JJeremy Anyone know why I'm getting 512 "Could not send your email." with the Mailchanne...

Chaika•11/6/23, 10:25 PM

You might want to reconsider mailchannels: workers-and-pages-discussions
Regardless, it needs to be your pages.dev (even if using a custom domain, has to be the pages.dev of your project), it corresponds with the cf-worker header. ex: cfid=myproject.pages.dev

CChaika You might want to reconsider mailchannels: https://discord.com/channels/59531799...

JeremyOP•11/6/23, 10:29 PM

Ah, figured. I don't think any of that is relevant if we have lockdown/SPF/DKIM though, right?

JeremyOP•11/6/23, 10:32 PM

Just wanted a simple webform

dXtee•11/8/23, 7:18 AM

Hi, looking for some help with functions to help me fetch json data from a third party API.

Apologies in advanced if I have got this all wrong, I am usually just front-end but I am helping this animal rescue organisation make their webpage - and to list the animals they have for adoption we are using a third party api that needs a API secret. To hide the API secret from the client-end I went about making an express.js backend that will be hosted and the end points will simply just call the
third party API with the secret, and some basic filtering queries will just be appeneded to the fetch.

As this is just a simple 'backend' I believe it would be best suited for serverless functions like on Cloudflare Pages Functions as I have migrated the static page to be hosted on Cloudflare Pages already. Would this implementation be possible in CloudFlare Pages functions. I tried looking through documentation and attempted to implement last night, would the basic functions act the same with a getRequest. I had issues with parsing the request for the queries as well as returning a proper response. An example query would be ".../listings?page=1&species=Cat"

If anyone has any examples they could point me to or any guidance I would greatly appreciate it.

Attached is an example endpoint for the express implementation.

alex (he/him)•11/8/23, 5:47 PM

Any plans to not bundle all functions into one worker and have a worker per function file? I understood that this is the case, but just learned recently that it is not. I really wish it would be a worker per function :/

kian•11/8/23, 5:50 PM

Not that I'm aware of - Workers (and Pages frameworks) projects become a single bundled Worker.

alex (he/him)•11/8/23, 7:17 PM

Yes, but it would be great if that could change in the future.

alex (he/him)•11/8/23, 7:17 PM

Getting the size limits per function file instead of the bundle

alex (he/him)•11/8/23, 7:17 PM

Being able to fine-tune each worker seperatly

alex (he/him)•11/8/23, 7:17 PM

Because if they get bundled, what are the actual benefits of using multiple files in functions folder over a single _workers file?

Aalex (he/him)Because if they get bundled, what are the actual benefits of using multiple file...

Hello, I’m Allie!•11/8/23, 7:35 PM

A file-based router, that you don't have to configure manually?

HHello, I’m Allie!A file-based router, that you don't have to configure manually?

alex (he/him)•11/8/23, 7:37 PM

That does my framework already, I'm just trying to understand some pros/cons here

Aalex (he/him)That does my framework already, I'm just trying to understand some pros/cons her...

Hello, I’m Allie!•11/8/23, 7:39 PM

You can also use Functions plugins, without having to rewrite them to match whatever system your framework uses

Hello, I’m Allie!•11/8/23, 7:39 PM

But yeah, from the perspective of a Framework author, I would generally go with a _worker.js_worker.js file

alex (he/him)•11/8/23, 7:42 PM

Thanks! That helps a lot.

HHello, I’m Allie!But yeah, from the perspective of a Framework author, I would generally go with ...

alex (he/him)•11/8/23, 7:42 PM

We @⁠Astro would love to see those functions being seperate workers, but yeah maybe a dream for a future update

Aalex (he/him)We @⁠Astro would love to see those functions being seperate workers, but yeah ma...

Hello, I’m Allie!•11/8/23, 7:44 PM

Is this to get around size limits, or just to solidify the limits per file?

alex (he/him)•11/8/23, 7:46 PM

Multiple reasons, size limits, custom config per page/route, separate api endpoints & frontend pages, better analytics, etc.

Aalex (he/him)Multiple reasons, size limits, custom config per page/route, separate api endpoi...

Hello, I’m Allie!•11/8/23, 7:57 PM

I do actually have an idea for how that could work, but it might make Pages more expensive

goldenhelix•11/9/23, 6:12 AM

I am using pages functions where my frontend react app (made with pages) talks to the server side in the functions. Although, this is working, but it is throwing the following message.

"Your worker created multiple branches of a single stream (for instance, by calling response.clone()response.clone() or request.clone()request.clone()) but did not read the body of both branches. This is wasteful, as it forces the system to buffer the entire stream of data in memory, rather than streaming it through. This may cause your worker to be unexpectedly terminated for going over the memory limit. If you only meant to copy the request or response headers and metadata (e.g. in order to be able to modify them), use the appropriate constructors instead (for instance, new Response(response.body, response)new Response(response.body, response), new Request(request)new Request(request), etc)."

I am not calling response.clone()response.clone() or request.clone()request.clone(). Can someone please explain what needs to be done in this case?

Ppetter I had the same warning, it disappeared when I upgraded to the latest version of ...

goldenhelix•11/9/23, 6:35 AM

Thank you! It worked!

micromashor•11/10/23, 1:10 AM

getting this error when trying to establish a logging session for a pages function on a preview deployment

TorontoPhysics•11/10/23, 2:05 PM

X-Forwarded-For gets appending when using functions:
I have a simple function attached to my page that runs fetch(). I use the request.headers.get("CF-Connecting-IP") to get the IP of the original requestor and get the 'X-Forwarded-For' header to that IP address. However, when the request goes out, the header is:
X-Forwarded-For: User-Ip, Cloudflare-IP

Where the IP of (I think) CF-Pages gets added to the X-Forwarded. This seems like odd behavior. While technically I believe the path is User -> CF Pages -> CF Function -> fetch-Request It seems this isn't really a double proxy with two Forwards as the CF Pages->CF Function I think should be transparent. In any case, is there a way to get a function in CF-Pages make a fetch with just the user's IP in the X-Forwarded-For header? If I don't manually add that header, CF sets the X-Forwarded-For header to just the Cloudfare-IP.

Rabid•11/11/23, 3:12 AM

Creating a simple link shortener with D1 and pages functions. Is there a way to programmatically create a cache rule once a shortened link is created? So subsequent requests to /myfunction/:id will return the cached 301 result instead of having to hit my D1 db and worker?

Oom Can I use cloudflare cache API in function? Also how to serve static assets with...

Hello, I’m Allie!•11/12/23, 11:16 AM

Yes to number one, don’t have any Functions on the route for number two

HHello, I’m Allie!Yes to number one, don’t have any Functions on the route for number two

Isaac McFadyen•11/13/23, 3:02 AM

Are routes.json auto-constructed now?

Isaac McFadyen•11/13/23, 3:02 AM

I thought you used to need to use a routes.json

Erisa•11/13/23, 3:51 AM

They are if you use /functions and have been for a long while (probably as long as theyve been a thing?)
You only need a custom one if youre doing something weird or are using a _worker.js

Poot•11/13/23, 11:53 AM

Greetings and salutations. I have a question about moving a Workers project in TypeScript over to Functions using advanced mode (with a _worker.js). I've written my application in the module syntax (as the docs instructed), but when I compile my TypeScript into _worker.js with tsc, in the output directory (which should be /build, as far as I can tell), the Pages project fails to deploy. I've tried using different options for "module" in the tsconfig.json as well, such as "amd" and "system", because "esnext" doesn't let me use the "outFile" option.

My question is: does all the code for a Function need to be compiled into a single _worker.js, or can each TypeScript file be compiled separately, and then the _worker.js will know to import from the other JavaScript files? Or would it be better to just give up on trying to migrate the Worker, and rebuild the application in the /functions folder instead?

PPoot Greetings and salutations. I have a question about moving a Workers project in T...

Hello, I’m Allie!•11/13/23, 12:02 PM

You can have a _worker.js_worker.js that just reexports your Worker. wranglerwrangler should autobundle for you anyway

Hello, I’m Allie!•11/13/23, 12:02 PM

You don't need to run tsctsc at all

Poot•11/13/23, 12:07 PM

Thank you for answering. So it would be as simple as copying over the /src folder and just adding a _worker.js with some minimal code that somehow exposes the Worker?
I'm not sure where I can find these steps in the docs, but the article about Advanced Mode said to compile all the TypeScript into a _worker.js, so I'm a little confused.

PPoot Thank you for answering. So it would be as simple as copying over the /src folde...

Hello, I’m Allie!•11/13/23, 12:18 PM

They added support a little while ago for _worker.js_worker.js bundling, so you could keep the same structure, and just have _worker.js_worker.js be

import worker from "../src";
export default worker;

import worker from "../src";
export default worker;

Poot•11/13/23, 12:41 PM

Oh, wow. Thank you! Going to give this a try.

Panda•11/14/23, 12:35 AM

@Isaac McFadyen | YYZ01 basically I just want the worker to check if the user is allowed to be on the site or not

Panda•11/14/23, 12:36 AM

if yes, allow them to load into it, if not, don't

Panda•11/14/23, 12:36 AM

the reason I want to do it with workers is because I don't want any site data/load to be sent before the user is verified

PPanda the reason I want to do it with workers is because I don't want any site data/lo...

Isaac McFadyen•11/14/23, 12:36 AM

What do you mean by sent?

Isaac McFadyen•11/14/23, 12:37 AM

Oh, as in, you want it to immediately block them without serving an HTML page?

Panda•11/14/23, 12:37 AM

yes

Isaac McFadyen•11/14/23, 12:37 AM

You should use SvelteKit middleware in this case, not _worker.js. SvelteKit already uses a _worker.js so you can't add your own.

Isaac McFadyen•11/14/23, 12:37 AM

Sorry, SvelteKit calls it Hooks not Middleware

Isaac McFadyen•11/14/23, 12:37 AM

You want the handlehandle hook: https://kit.svelte.dev/docs/hooks#server-hooks-handle

Panda•11/14/23, 12:38 AM

Oh. But I mostly wanted to make it using a worker since I want to use this in other projects as well, guess this will work though

Isaac McFadyen•11/14/23, 12:38 AM

Basically, check their auth (cookie or header or whatever) and if it's invalid just don't call resolveresolve, return an error instead.

alex (he/him)•11/14/23, 4:24 PM

Would love to get some insights here: https://canary.discord.com/channels/595317990191398933/1174021841023664288
I'm currently trying to fix _routes.json_routes.json generation in Astro's cloudflare adapter

Anyone know why I'm getting 512 "Could not send your email." with the Mailchannels plugin? I believe

Similar Threads

Similar Threads

Similar Threads