So what's the point of this code without a DO? https://discord.com/channels/595317990191398933/77939

LLordSilver So what's the point of this code without a DO? https://discord.com/channels/5953...

C

Chaika•9/20/24, 3:28 PM

you get a session/browser instance/puppeteer instance you can use

L

LordSilverOP•9/20/24, 3:28 PM

But you would get it even without the keepalive

C

Chaika•9/20/24, 3:29 PM

the keepalive is useless in that case because you'd never be reusing the session

C

Chaika•9/20/24, 3:29 PM

If you browser.close()'d, it'd go away. If you just disconnected or died, you'd just have an orphaned session

L

LordSilverOP•9/20/24, 3:29 PM

So you are saying the puppeteer machine would kill the browser even if you specifically told it to keep the browser open for 10m?

LLordSilver So you are saying the puppeteer machine would kill the browser even if you speci...

C

Chaika•9/20/24, 3:30 PM

no but it'd be useless to you lol

L

LordSilverOP•9/20/24, 3:30 PM

Yeah I know, that's what I was saying.

LLordSilver If I use this code, but I don't have any DO, what happens?

C

Chaika•9/20/24, 3:30 PM

if you run this over and over you'd just keep getting new

L

LordSilverOP•9/20/24, 3:30 PM

So the browser keeps running for 10m, but you have no way to reconnect to the same session.

L

LordSilverOP•9/20/24, 3:30 PM

It's more like wasted resources for Cloudflare.

LLordSilver So the browser keeps running for 10m, but you have no way to reconnect to the sa...

C

Chaika•9/20/24, 3:30 PM

You do have a way to reconnect, you're not just using it in that example

C

Chaika•9/20/24, 3:30 PM

see https://developers.cloudflare.com/browser-rendering/get-started/reuse-sessions/ for how to reuse sessions

Cloudflare Docs

Reuse sessions | Browser Rendering docs

The best way to improve the performance of your browser rendering worker is to reuse sessions. One way to do that is via Durable Objects, which allows you to keep a long running connection from a worker to a browser. Another way is to keep the browser open after you’ve finished with it, and connect to that session each time you have a new reques...

L

LordSilverOP•9/20/24, 3:31 PM

Yeah I mean with a DO.

C

Chaika•9/20/24, 3:31 PM

You don't need a DO to reuse sessions although one session is tied to one execution so limited in usage vs a DO can just keep processing/queue

C

Chaika•9/20/24, 3:32 PM

side note: I agree that all that manual management is silly and in a serverless platform CF should really just do all that for you and let you just use the api without worrying about sessions or instances. We've brought that up to them before, and the session APIs were a bit of their response to it, still annoying you have to manually manage
Using the DO to reuse your session and the sessions API are two entirely separate ways of extending the life of a session/browser instance to reduce cold starts/limits

L

LordSilverOP•9/20/24, 5:10 PM

What sessions API?

L

LordSilverOP•9/20/24, 5:12 PM

But yeah, this looks way more complex than just being able to run puppeteer directly on your worker. If you are making high-level abstraction, at least make them right so that there's a good DX that truly hides stuff happening behind the scenes.

P

Patrick M•9/20/24, 5:14 PM

With

nodejs_compat_v2

nodejs_compat_v2

nodejs_compat_v2

nodejs_compat_v2 you can use

nodemailer

nodemailer

nodemailer

nodemailer now! SMTP transport does not work but you can import the library, build an email payload, and probably use other transports like to Amazon SES too. Example | Source

LLordSilver What sessions API?

C

Chaika•9/20/24, 5:45 PM

You can do

puppeteer.sessions(env.YourBrowserBinding)

puppeteer.sessions(env.YourBrowserBinding)

puppeteer.sessions(env.YourBrowserBinding)

puppeteer.sessions(env.YourBrowserBinding) like in this example here https://developers.cloudflare.com/browser-rendering/get-started/reuse-sessions/

Cloudflare Docs

Reuse sessions | Browser Rendering docs

The best way to improve the performance of your browser rendering worker is to reuse sessions. One way to do that is via Durable Objects, which allows you to keep a long running connection from a worker to a browser. Another way is to keep the browser open after you’ve finished with it, and connect to that session each time you have a new reques...

C

Chaika•9/20/24, 5:46 PM

then you can filter out active sessions, grab the session id, and .connect rather then .launch to reuse it

L

LordSilverOP•9/20/24, 6:10 PM

Why it has to be so complicated.

D

Dean•9/20/24, 9:56 PM

anyone else notice you can't Deploy through CF workers via the dashboard? If you modify any code, the Deploy button remains disabled.

O

orangeboi•9/21/24, 1:49 AM

i'm getting this error all of a sudden in my deploys and I can't explain it.

I can run npm run dev, npm run build, npm run serve no problem

N

Name With Exactly 32 Characters!•9/21/24, 6:35 AM

do i have a skill issue or is this not a valid cron

[triggers]
crons = ["0 0 * * 0"] # Weekly trigger

[triggers]
crons = ["0 0 * * 0"] # Weekly trigger

[triggers]
crons = ["0 0 * * 0"] # Weekly trigger

[triggers]
crons = ["0 0 * * 0"] # Weekly trigger

X [ERROR] A request to the Cloudflare API failed
  invalid cron string [code: 10100]

X [ERROR] A request to the Cloudflare API failed
  invalid cron string [code: 10100]

X [ERROR] A request to the Cloudflare API failed
  invalid cron string [code: 10100]

X [ERROR] A request to the Cloudflare API failed
  invalid cron string [code: 10100]

R

raj.js•9/21/24, 5:50 PM

Why is

new Date().toString()

new Date().toString()

new Date().toString()

new Date().toString()

giving Thu Jan 01 1970 00:00:00 GMT+0000 (Coordinated Universal Time)
instead of current time, when run on production? it gives correct curent time when worker run locally

Rraj.js Why is ```new Date().toString()``` giving Thu Jan 01 1970 00:00:00 GMT+0000 (Co...

A

Ashley•9/21/24, 6:12 PM

Are you executing that in the global scope by any chance, rather than in the event handler?

J

jam_zaw•9/22/24, 4:39 AM

Hello everyone,
I'm looking to implement a caching strategy for my Cloudflare Worker to optimize the API responses from my database. Would leveraging the default Cloudflare Cache be sufficient for this use case, or would it be more effective to use an external tool like Upstash? Perhaps even a combination of both?
Any insights or recommendations would be appreciated!

W

Walshy•9/22/24, 11:35 AM

Let me report to the team

W

Walshy•9/22/24, 11:38 AM

That is indeed a very old binding

N

Nelums•9/22/24, 10:14 PM

Sorry for the question, I was searching for help respect a problem I have with my IP and it seems that Cloudflare is involved, is there a channel where I can get any help with it? or it is mandatory to send an email to the company?

N

Nelums•9/22/24, 10:15 PM

You read my mind sir thanks you

AAndré Hey all, I'd love to learn a bit about the design space of the worker api. Any c...

N

NoGameNoFun•9/23/24, 4:32 AM

what is the status for this synchronous XHRs?

L

Leo_C•9/23/24, 8:31 AM

wired bug, is there a problem with the validation regex???

L

Leo_C•9/23/24, 8:31 AM

L

Leo_C•9/23/24, 8:31 AM

L

Leo_C•9/23/24, 8:34 AM

L

Leo_C•9/23/24, 8:38 AM

This is the UI for worker to add routes.

L

Leo_C•9/23/24, 8:46 AM

L

Leo_C•9/23/24, 9:28 AM

Yes, I can create this route using the API, but not on the UI.

L

Leo_C•9/23/24, 9:57 AM

Odd lengths are allowed, but not even lengths.

A

Ashley•9/23/24, 1:46 PM

Anyone using email workers and seen the catch all status just hang? I set it up just now, and it's "flashing" (the status)... only network request I can see that failed was a 404 to zones/subscriptions but not sure if that'd be used for this.

A

Ashley•9/23/24, 1:47 PM

(If I send emails, they don't bounce, but they also don't trigger my worker)

A

Ashley•9/23/24, 1:50 PM

Ah one did just bounce, but not all of them curiously - the email routing status page says it's synced so it should be routing, but I presume the catch all isn't working due to the flashing loading indicator on the status

A

Ashley•9/23/24, 1:51 PM

Yeah it looks bugged

This isn't the real domain I'm using, but I added a second route and it's just hanging on retrieving the status - the 404 exists on this page, not sure if it's the reason, and it's when it tries to call

https://dash.cloudflare.com/api/v4/zones/xxx/subscription

https://dash.cloudflare.com/api/v4/zones/xxx/subscription

https://dash.cloudflare.com/api/v4/zones/xxx/subscription

https://dash.cloudflare.com/api/v4/zones/xxx/subscription

R

Raymond Camden•9/23/24, 3:05 PM

hey folks. i posted to workers-help, but haven't gotten well, help yet

so trying here. I've got a worker that uses a schedule. locally, it runs just fine. but in production, something weird happens. when it tries to use http to hit an API, it gets a 521 response. now here's the weird thing. i added a fetch() to the code so i could run my worker via the browser. the same network call there works, so its literally just failing when cloudflare executes it on a schedule. it's a bit messy now, but the code is here -https://github.com/cfjedimaster/cloudflareworkers-demos/blob/main/getincidents/src/index.js any ideas?

GitHub

cloudflareworkers-demos/getincidents/src/index.js at main · cfjedim...

Repository for CF Workers demos/examples/etc. Contribute to cfjedimaster/cloudflareworkers-demos development by creating an account on GitHub.

A

azizthewhiz•9/23/24, 3:27 PM

I posted how to update Cloudflare Workers project secrets on git push in my personal LinkedIn and X but figured folks here might benefit as well.

If you’re tired (like I was) of copy-pasting environment secrets and setting them manually via dashboard or wrangler CLI I wrote a git pre-push script that automates that.

Set CL secrets in
```
.env.cloudflare
```
.env.cloudflare
```
.env.cloudflare
```
.env.cloudflare

Create

CLOUDFLARE_API_TOKEN

CLOUDFLARE_API_TOKEN

and add it to

.env.production

.env.production

.env.production

.env.production

Create
```
env-to-json.sh
```
env-to-json.sh
```
env-to-json.sh
```
env-to-json.sh
that generates
```
secrets.json
```
secrets.json
```
secrets.json
```
secrets.json
Update
```
pre-push
```
pre-push
```
pre-push
```
pre-push
git hook

Now your secrets will update on every git push - no more forgetful copy-pasta.

P.S. #3 is optional if you don't mind storing CL secrets in

secrets.json

secrets.json

secrets.json

secrets.json

C

Corsix👾•9/23/24, 8:47 PM

to have several service bindings to other workers in one worker, what's the correct way of formatting it in toml? nested array? like this?

services = [
  { binding = "WORKER_A", service = "worker-a" }, 
  { binding = "WORKER_B", service = "worker-b" }
]

services = [
  { binding = "WORKER_A", service = "worker-a" }, 
  { binding = "WORKER_B", service = "worker-b" }
]

services = [
  { binding = "WORKER_A", service = "worker-a" }, 
  { binding = "WORKER_B", service = "worker-b" }
]

services = [
  { binding = "WORKER_A", service = "worker-a" }, 
  { binding = "WORKER_B", service = "worker-b" }
]

So what's the point of this code without a DO? https://discord.com/channels/595317990191398933/77939

Similar Threads

Similar Threads

Similar Threads