Hey! Just curious, I have a .env file with my secrets, and then they're in CF Workers for my environ

Hey! Just curious, I have a .env file with my secrets, and then they're in CF Workers for my environment, but when I run preview with the workers, it doesn't seem to load the .env file, and there's no docs on how to "declare" a var in the wrangler.json that's a secret

ZZachHandley Hey! Just curious, I have a .env file with my secrets, and then they're in CF Wo...

Hard@Work•11/5/25, 11:12 PM

When you "run preview", do you mean like

wrangler dev --remote

wrangler dev --remote

ZachHandleyOP•11/5/25, 11:12 PM

nvm so I can wrangler dev, can I wrangler preview?

ZachHandleyOP•11/5/25, 11:12 PM

I am trying to preview so like, my local build is fine, and when I run wrangler dev it is fine, but when I go to my site, there's an error

ZZachHandley nvm so I can wrangler dev, can I wrangler preview?

Hard@Work•11/5/25, 11:12 PM

IIRC they got rid of wrangler previewwrangler preview?

ZachHandleyOP•11/5/25, 11:13 PM

one complaint I have is that it's SO hard to find these weird errors that occasionally happen, even with observability

ZachHandleyOP•11/5/25, 11:13 PM

when the workers build has an error that dev doesn't

ALECKS•11/6/25, 2:17 AM

Man, for all the things I love about Cloudflare, I keep running into some very painful friction points.

Cannot access "WSTenantRouter#publish" as Durable Object RPC is not yet supported between multiple dev sessions.

... so frustrating.

What are we supposed to do for intra worker/DO messaging? Queues don't work when using multiple dev commands (which you have to do when using Vite). Now I see that DO's can't make RPC calls across dev sessions? Workers can, why can't DOs?

--edit: Looks like .fetch() works from DO to worker in another dev session

Samantas5855•11/6/25, 10:43 AM

is something wrong with workers?

Samantas5855•11/6/25, 10:43 AM

My api pages never load

Samantas5855•11/6/25, 10:43 AM

they just load forever until they timeout

Samantas5855•11/6/25, 10:48 AM

only the ones using d1

AnotherAlex•11/6/25, 3:38 PM

Hi, has anyone used the placeholders functionality in cloudflare pages _redirects file? I'm getting the same error as this individual but I also can't find any examples except for the official docs.
https://community.cloudflare.com/t/splats-or-placeholders-parameters-not-work-in-redirects-file/492924

johtso•11/6/25, 6:23 PM

is it possible to configure builds through the wrangler config file?

johtso•11/6/25, 6:24 PM

I'm going to have multiple envs for my worker and I'd rather not have to go into the dashboard and manually configure automatic builds for each one

texan•11/6/25, 9:01 PM

Not possible today, no

rbm•11/6/25, 11:52 PM

Checking before I go to -help -- is there a pattern for preserving old assets or implementing "skew protection" on CF Workers + Static Assets that I'm missing? It doesn't seem like there's any complete solution for this on the current version of Workers that I can find.

Context: we're running into the issue where we frequently deploy an SPA to Workers with Static Assets and users who are on stale versions of the app encounter 404s when assets lazy-load and it breaks the app., The assets have a fingerprint in them (produced by Vite) and it changes between deploys. The old assets / fingerprints are immediately invalidated on deploy (AFAICT.) This is a pure SPA that doesn't use SSR at all, there is a Workers script that proxies our API that is invoked when no asset matches; it passes API paths to the backend API service and otherwise just fetches from env.ASSETSenv.ASSETS.

Rrbm Checking before I go to -help -- is there a pattern for preserving old assets or...

Hard@Work•11/6/25, 11:54 PM

https://developers.cloudflare.com/workers/static-assets/routing/advanced/gradual-rollouts/#preventing-asset-mismatches-with-version-affinity

Cloudflare Docs

Gradual rollouts

Provide static asset routing solutions for gradual Worker deployments.

HHard@Work https://developers.cloudflare.com/workers/static-assets/routing/advanced/gradual...

rbm•11/6/25, 11:59 PM

I've seen this, but our session IDs are actually persistent across full page reloads, so it seems like users would get "stuck" on an old version until their session recycled (which could be a long time in this case) and the other solutions I can think of to key the affinity (e.g. set a cookie with a nonce on initial document load) I think would break when there's multiple tabs / etc.

zapalona_swieczka•11/7/25, 8:12 AM

hey, hm, do I underatand correctly, that cache purge is available to non enterprise, but to save cache with tag one need enterprise acount?

Sstavros-k Hi! I'm trying to increase the project (pages) limits (100). Tried both the form...

stavros-k•11/7/25, 8:45 AM

Hello, Can someone give me pointers on how to proceed with this?

SAKAKOI•11/7/25, 8:49 AM

I added a my website on cloudflare workers and and deleted old DNS and Cname for the domain. I can access the website via url and its working fine but google giving error that DNS error. What settings or records do I need to update. Kindly help

SSAKAKOI I added a my website on cloudflare workers and and deleted old DNS and Cname for...

Hello, I’m Allie!•11/7/25, 2:26 PM

Might just be DNS propagation lag? Give it a bit of time, and then check back

VJ•11/7/25, 3:28 PM

Hi , I am Vijayabaskar from India, I am building a TANCN (OSS Project), I have my application deployed on Cloudflare, I have also purchased a domain tancn.dev from the cloudflare itslef, I love cloudfare and wrangler in general ,

Now I have a issue which is been bugging me lately, that is my site tancn.dev is randomly throwing 500 error or This Site Can't be reached error from the user

Here is a issue from one of my users : https://github.com/Vijayabaskar56/tancn/issues/30 , and I am not sure that is affecting certain region or affecting randomly , there is no error reported on the dashboard or misconfigured domain

here is some other user also reporting
https://x.com/IndieCEO/status/1985284378784563654
https://x.com/daveycodez/status/1984830994906824960

is anyother people facing this issues , or know why is this happening and how to resolve this issues

GitHub

Website Connection failing · Issue #30 · Vijayabaskar56/tancn

I am in the USA, using normal USA WiFi networks. Whenever I try to access the website link, there is a connection error. I tried on multiple different WiFi networks and devices and still get the sa...

Indie CEO ⚡️ (@IndieCEO)

@vijayabaskar56 domain not working

•

11/3/25, 9:54 AM

daveycodez (@daveycodez)

@vijayabaskar56 500 rn jsyk

•

11/2/25, 3:52 AM

VVJ Hi , I am Vijayabaskar from India, I am building a TANCN (OSS Project), I have m...

VJ•11/7/25, 3:30 PM

my codebase itslef opencode , if there is any misconfiguration also please let me know , I wanetd this to be resolve asap , as I don't wanted my user to experience this

Original message was deleted

Hard@Work•11/7/25, 4:57 PM

Not anything documented: https://developers.cloudflare.com/api/resources/pages/

The dashboard does support single-file uploads iirc, so maybe you can use that?

Cloudflare API | Pages

Interact with Cloudflare's products and services via the Cloudflare API

nico•11/7/25, 6:08 PM

Hi, I am currently trying to migrate one of my Cloudflare Pages Nuxt (+ Nitro backend) applications to Cloudflare workers but I can't find the equivalent for the environment seperation (prod vars vs preview vars). I looked a bit around in the configuration and found a documentation that looked promising but is putting secrets into my Cloudflare Workers environment really only possible through the Wrangler CLI as described in this documentation? https://developers.cloudflare.com/workers/wrangler/environments/#_top

Until now I have just deployed through the Git integration and haven't really looked into the Cloudflare CLI tools since it worked out for my 10+ Pages Nuxt apps perfectly (i dont even know if Pages has an equivalent for that)

Any tips or guidance would be much appreciated, thank you very much (ui-way to add secrets for specific envs? only possible through the cli?)

Nnico Hi, I am currently trying to migrate one of my Cloudflare Pages Nuxt (+ Nitro ba...

texan•11/7/25, 7:56 PM

it’s not you, it’s us.

Environments have been a huge topic of discussion. There’s lot of info in discord if you search around in help threads.

I would suggest getting started by configuring workers environments and WCI. From there, many help threads discussing how to configure the environments with WCI. We might need a full walk through in our docs about how to set this up fully though

Ttexan 😭 it’s not you, it’s us. Environments have been a huge topic of discussion. Th...

nico•11/7/25, 8:08 PM

Ahh alright, good to know, at least I’m not just blind then

Is this separation planned to be visible / configurable in the frontend (Workers UI) at some point? Just wondering if it’s worth waiting a few months before migrating, or if there’s no information on that yet.

WCI is not a common term for me, does that basically just mean to integrate this Worker CLI (it is called wrangler i guess) into my CI (GitHub Actions in my case)?

Thanks again for the super quick reply, that already helps a lot

texan•11/7/25, 8:10 PM

Ah, that’s shorthand for Workers CI (builds). There were some echos on making this experience better via dash.

Wrangler environments is a way to configure multiple workers with separate bindings, using the same underlying codebase. You can use this concept to create a production worker and staging worker. They’d be separate workers, but you can definitely set up workers CI to automatically upload, deploy, etc to those environments based on branch

imightbekyle•11/8/25, 2:45 AM

Cloudflare removing some worker environment variables added in console after deploying is wild.
Any tips here?

Iimightbekyle Cloudflare removing some worker environment variables added in console after dep...

1984 Ford Laser•11/8/25, 3:00 AM

Did you modify in the dash after deploying in console?

11984 Ford Laser Did you modify in the dash after deploying in console?

imightbekyle•11/8/25, 3:01 AM

yes

Iimightbekyle yes

1984 Ford Laser•11/8/25, 3:02 AM

Could you see those variables in the dash?

11984 Ford Laser Could you see those variables in the dash?

imightbekyle•11/8/25, 3:03 AM

I could see them yes, but not after ci pipeline deploys

mnathl•11/8/25, 3:07 AM

Hey guys if I'm having an issue with a cloudflare error, would I be able to ask for it in this chat? I posted something on the forum, but wasn't sure if I would be able to ask for help here.

Iimightbekyle Cloudflare removing some worker environment variables added in console after dep...

texan•11/8/25, 3:34 AM

Recommend checking out https://developers.cloudflare.com/workers/wrangler/configuration/#source-of-truth

Cloudflare Docs

Configuration

Use a configuration file to customize the development and deployment setup for your Worker project and other Developer Platform products.

Mmnathl Hey guys if I'm having an issue with a cloudflare error, would I be able to ask ...

texan•11/8/25, 3:35 AM

#workers-and-pages-help thread so it doesn’t get buried (or just ask here!)

Ttexan #workers-and-pages-help thread so it doesn’t get buried (or just ask here!)

mnathl•11/8/25, 4:30 AM

Sweet. Yea I asked there before. Basically I'm running into this issue (out of nowhere) where some API requests on my NextJS backend are returning this Cloudflare error:

For context, I'm trying to access my MongoDB collection; however it seems to also happen with other API calls as well. I don't really know where it's come from, it started out of nowhere and the documentation isn't super helpful.

Dubz•11/8/25, 8:54 AM

So I have a genuine question. I know that workers are limited to 100k executions per day (on the free plan), and this is shared amongst all account workers, not per-worker. However, there's not really anything preventing users from just making another Cloudflare account to get another 100k executions. Obviously there will be limitations as data can't be shared easily, but if you're making multiple, unrelated projects, that wouldn't matter. Even if a domain is needed on your account, you can get super cheap ones for like $4/yr (which is less than one month of pro plan). Has this ever been considered and thought to change? It'd be cool if we could create "projects" where they have clusters of various services and each of those have their own pool of usage (although I hope that doesn't change them to lower thresholds, in which case forget this entire message lol).

VVJ Hi , I am Vijayabaskar from India, I am building a TANCN (OSS Project), I have m...

VJ•11/8/25, 9:46 AM

anyone ? know why is this happening
Here i some additional info , when this happen this is the log that I getting to the user
{"status":500,"unhandled":true,"message":"HTTPError"}

Original message was deleted

Dubz•11/8/25, 10:22 AM

If there are legitimate separate purposes though then not entirely. Companies segregate parts of their business under different names for many reasons. That would be separate entities then so not really applicable, and they could realistically spread put whatever they're trying to do with that. Of course there's probably gonna be some person to take things to the extreme with like 50+ accounts just to abuse it and ruin it for everyone else. I only separate mine for business/personal stuff because separation of access is just good practice.

Borna Punda•11/8/25, 1:06 PM

Hey, so per-branch/commit previews sound great, but I see one major drawback. Couldn't someone just make a commit that dumps all the environment secrets and use that exfiltrate them? Now, I know only trusted people should have push access, but that is still a privilege escalation.

Borna Punda•11/8/25, 1:06 PM

Or do previews not get access to production secrets?

DDubz So I have a genuine question. I know that workers are limited to 100k executions...

PhoenixSpace•11/8/25, 1:54 PM

This honestly seems like a massive pain to avoid paying like $5/month

Original message was deleted

Borna Punda•11/8/25, 3:16 PM

There are quite a few steps standing between a feature branch and production on well-structured repos

Borna Punda•11/8/25, 3:16 PM

Including review processes, CI pipelines, and whatnot

Borna Punda•11/8/25, 3:22 PM

But that aside, we figured out how to set non-production builds to a different environment from the dashboard

PPhoenixSpace This honestly seems like a massive pain to avoid paying like $5/month

Dubz•11/8/25, 4:29 PM

Honestly you aren't wrong. I only really see kids doing this because they have no money lol

mikerowave 🐈•11/9/25, 6:07 AM

just bought a new domain from CF and set it as the domain for a workers project
do i need to do anything else or should i just wait for the domain to work for the project? (ive already rebuilt the project)

Hey! Just curious, I have a .env file with my secrets, and then they're in CF Workers for my environ

Similar Threads

Similar Threads

Similar Threads