is anyone else having issues with executing discord webhooks from workers

TrueBrain•3/9/24, 2:11 PM

You are not alone. Since a few hours, Discord Webhook calls via Cloudflare Workers return a 403 for me, with some internal error message. If I would guess, Discord added something to their block-list for one reason or another

Kai•3/9/24, 2:59 PM

Experiencing same issue with Discord Webhook calls

AAnthony is anyone else having issues with executing discord webhooks from workers

TrueBrain•3/9/24, 3:57 PM

and it is working for me again

TTrueBrain and it is working for me again 🙂

AnthonyOP•3/9/24, 4:14 PM

Same, Great!

dave•3/9/24, 6:57 PM

when doing a subrequest fetch to a html file, why does the etag and content-length header disappear?

dave•3/9/24, 7:05 PM

ah, I see I needed to have no-transform on the asset's cache-control header

dave•3/9/24, 7:23 PM

that's annoying, even if I set

no-transform

no-transform

on the request, it still doesn't work..

phyx•3/10/24, 4:22 AM

Has anyone tried submitting custom metrics from your worker to datadog via api? Without an agent, it's pretty hard to batch send a bunch of metrics because all you can do is append an array of metrics and "sendAll" with ctx.waitUntil() after the request event is over. Was wondering if anyone have tried this and how did it affect your DD bill?

Walshy•3/10/24, 7:05 AM

So this is referring to cold start, it used to be slower for free plan on first request. These days, every worker regardless of plan will be fast on first and subsequent requests

Walshy•3/10/24, 7:51 AM

They aren't always running on every single metal (since we reserve memory for each one) but yes, they can run on every single one and starting an isolate is lightning fast

At CF, we believe on running every service in every colo on every metal. We have a few these days which aren't but it's more work to not do that than do it so we will always strive to do that

Walshy•3/10/24, 7:51 AM

It'd be slower to go to another Colo than just spin up on the initial one so we just spin up and serve!

Walshy•3/10/24, 7:52 AM

These are isolates not VMs or anything so there's barely any startup at all, it's very fast and that's why we chose this model

Walshy•3/10/24, 7:56 AM

Yep yep and yep! The only difference these days in that kind of regard is the amount of scripts and script sizes

Walshy•3/10/24, 7:57 AM

But we're continuing to raise both of those limits all the time as we see people using the platform more

Walshy•3/10/24, 7:59 AM

Some more info:
* https://blog.cloudflare.com/cloud-computing-without-containers
* https://developers.cloudflare.com/workers/reference/how-workers-works/
* https://developers.cloudflare.com/workers/reference/security-model/

Walshy•3/10/24, 8:09 AM

Yeah so that's the time waiting before the request is sent. Could be a number of things like concurrency limits. If you have a lot of other ongoing requests, it may be waiting for those to finish

You may be able to find out more info with a net log - https://www.chromium.org/for-testers/providing-network-details/

WWalshy Some more info: * <https://blog.cloudflare.com/cloud-computing-without-container...

BlasterLord•3/10/24, 8:10 AM

The first article is really informative. Pretty cool stuff

Walshy•3/10/24, 8:13 AM

If you capture a netlog of the issue, feel free to toss it my way, I doubt it's anything related to us but always good to check anyway

You can email it over

walshy@

walshy@

SageSphinx63920•3/10/24, 2:07 PM

Anyone has some problems with emails not being sent within a worker?

SageSphinx63920•3/10/24, 2:08 PM

I have some problem with the promise never getting executed

Deo Kumar•3/10/24, 2:31 PM

As Workers sites uses kv, do the storage and all the read operations count in account?

SageSphinx63920•3/10/24, 2:33 PM

Asaik, yes

SSageSphinx63920 Asaik, yes

Deo Kumar•3/10/24, 2:40 PM

Why does it always show 0B even though I have several workers site kv namespaces with assets in it.

SageSphinx63920•3/10/24, 3:00 PM

And there are kv namespaces shown on the left part of the site?

cora•3/10/24, 5:01 PM

trying to setup upstash, where is the integrations tab in the dashboard?

Ccora trying to setup upstash, where is the integrations tab in the dashboard?

Chaika•3/10/24, 6:35 PM

Only Workers have those integrations, under the integrations tab. It looks like you are in a Pages Project

DDeo Kumar As Workers sites uses kv, do the storage and all the read operations count in ac...

Chaika•3/10/24, 6:37 PM

If you use Cloudflare Pages, you get unlimited static asset requests.
The 0B is weird though, there's some sampling and slowness on updates. Like Sage was asking, you see the namespaces in that account?

CChaika If you use Cloudflare Pages, you get unlimited static asset requests. The 0B is...

Deo Kumar•3/10/24, 7:01 PM

Yes, the namespaces are showing but I am seeing the 0B usage since I created the account. It never updates, only the read/write matrics update.

Deleted User 7f983ae1•3/11/24, 12:10 AM

Just out of curiosity, will durable object be ever released down to free plan, if anyone know

Thenerox•3/11/24, 3:07 AM

Hello one question its there any possible way to block access to my workers, example make them only accessible via AWS APIM or Azure APIM? and if i try to access the workers.dev url directly from browser its blocked

Deleted User 7f983ae1•3/11/24, 8:03 AM

Yes, by implement so only certain IP block can access it and any other return a 4xx error

Advany•3/11/24, 12:54 PM

I have a wierd requests by a customer. They want all requests from a worker to be served from one country. Is that possible (on a non ent plan)? or Do I need to use a non serverless solution?

AAdvany I have a wierd requests by a customer. They want all requests from a worker to b...

Hello, I’m Allie!•3/11/24, 12:56 PM

Sounds like https://developers.cloudflare.com/data-localization/regional-services/ maybe?

Cloudflare Docs

Regional Services · Data Localization Suite docs

Regional Services gives you the ability to accommodate regional restrictions by choosing which subset of data centers decrypt and service HTTPS …

dave•3/11/24, 3:03 PM

I'm getting this error on one of my wrangler tailwrangler tail commands:

Successfully logged in.
Expired! Auth code or refresh token needs to be renewed.

Successfully logged in.
Expired! Auth code or refresh token needs to be renewed.

dave•3/11/24, 3:03 PM

it instantly expires

captain_enjoyable•3/11/24, 3:16 PM

I see workers-and-pages-discussions about this previously but am still pretty confused.

I want to want to clear worker cache globally for a given request path e.g. www.myexample.com/123www.myexample.com/123 . I have populated the cache by doing cache.put()cache.put() It seems very clear that the worker API cache.delete()cache.delete() will only remove the item from that colo's cache. Is there a way to purge the item from all worker's caches in all colos?

The thread linked seems to indicate I can do so using the API.

Ccaptain_enjoyable I see [some discussion](https://discord.com/channels/595317990191398933/77939007...

Unsmart•3/11/24, 3:50 PM

The docs explain this here: https://developers.cloudflare.com/workers/runtime-apis/cache/#delete

The cache.delete method only purges content of the cache in the data center that the Worker was invoked. For global purges, refer to Purging assets stored with the Cache API.

Nicolas•3/11/24, 4:07 PM

Are there any thoughts on creating an Auth Worker for fullstack developers to use? Similar to AWS Cognito?

I am currently developing a Cloudflare worker that leverages D1 and Lucia to implement this, and it works for sign-up/sign-in, but if there was already Cloudflare service that could handle this, that would be awesome! Because i still need to implement auth workflows like forgot/reset password, email sign-in, magic links, Google OAuth, etc.

UUnsmart The docs explain this here: <https://developers.cloudflare.com/workers/runtime-a...

captain_enjoyable•3/11/24, 5:02 PM

I'm still confused. The doc you linked says for Global Purges visit here when I visit there it says.

To purge an asset globally, you must use the standard cache purge options. Based on cache API implementation, not all cache purge endpoints function for purging assets stored by the Cache API.

and also

Currently, it is not possible to purge a URL stored through Cache API that uses a custom cache key set by a Worker. Instead, use a custom key created by Page Rules. Alternatively, purge your assets using purge everything, purge by tag, purge by host or purge by prefix.

This does not make it particularly clear how the worker caches.defaultcaches.default interacts with the global API.

What I'm gathering is because "not all cache purge endpoints function for purging assets stored by the Cache API" my only option is to use the global purge or upgrade to enterprise and use cache tags. Is that right?

NNicolas Are there any thoughts on creating an Auth Worker for fullstack developers to us...

dave•3/11/24, 5:13 PM

my company made this internally for ourselves, we literally have a project called auth-workerauth-worker internally

NNicolas Are there any thoughts on creating an Auth Worker for fullstack developers to us...

dave•3/11/24, 5:13 PM

There's only Cloudflare Access, but the pricing does not scale at all for SaaS use-cases.

is anyone else having issues with executing discord webhooks from workers

Similar Threads

Similar Threads

Similar Threads