Cloudflare Developers•2y ago

Is there any sort of "on exit" or "on close" event callback for workers?

limikael•3/15/24, 7:04 AM

What has happened to the "Triggers" tab?

limikael•3/15/24, 7:04 AM

There used to be a "Triggers" tab, right? To set up a custom domain for the worker... Or am I dreaming?

JJoeDoe Is there any sort of "on exit" or "on close" event callback for workers?

Hello, I’m Allie!•3/15/24, 7:14 AM

There is not

Llimikael What has happened to the "Triggers" tab?

Hello, I’m Allie!•3/15/24, 7:15 AM

It’s here: https://dash.cloudflare.com/?to=/:account/workers/services/view/:worker/production/settings/triggers

andifined.dev•3/15/24, 7:25 AM

I have a question regarding the design of worker logics. I frequently encounter situations where I want to extract the logic required by a fetch handler into separate functions. However, as I proceed with the extraction steps, I find myself needing access to the request and environment objects within nested functions. I'm struggling to find a solution for designing these parent-child relationships without having to pass the request, context, and environment objects several levels deep. Is there a solution for this issue?

HHello, I’m Allie!It’s here: https://dash.cloudflare.com/?to=/:account/workers/services/view/:work...

limikael•3/15/24, 7:30 AM

Ah ok, and the way to reach them in the UI is via Settings >> Triggers. But just for the record, this has changed, hasn't it? It used to be directly on root level (same tab row as Setting?). Just checking if I'm geting dementia...

Aandifined.dev I have a question regarding the design of worker logics. I frequently encounter ...

limikael•3/15/24, 7:33 AM

I would say that passing it down is fine. This is also the same as other platforms, e.g. Node.js. If you don't want to pass down three objects, you could create your own class to wrap request, env and context, and pass down instances of that class.

Llimikael Ah ok, and the way to reach them in the UI is via Settings >> Triggers. But just...

Hello, I’m Allie!•3/15/24, 7:34 AM

Yes, there was a redesign recently

HHello, I’m Allie!Yes, there was a redesign recently

limikael•3/15/24, 7:39 AM

phew... good to know...

jepp•3/15/24, 7:48 AM

Good Morning, if I may ask, I have a question regarding a worker prototype I've recently built. Essentially, it functions as a bundled worker serving as a WebSocket server

[yes!]. It listens for incoming parameterized requests, looks up a value in KV, and returns the value via the WebSocket connection. Due to the inherent behavior of bundled workers having a 50 ms lifetime, the WebSocket connection is automatically closed or terminated. This behavior is intentional. I utilize the WebSocket connection to transmit data to the client in a manner that is intended to make data scraping (e.g., via REST) more challenging.

Now, here are my questions: How does a worker, in general, scale concerning WebSockets? Or is this a general inquiry regarding the behavior of bundled workers concerning concurrent connections? I'd like to determine how many parallel WebSocket connections/requests I could potentially serve, both in the free plan and the smallest paid plan initially. Additionally, I don't yet grasp whether it's my responsibility as a developer to spin up the appropriate number of workers myself or if this process is handled automatically by the Cloudflare infrastructure (meaning only one worker needs to be programmed, and the rest occurs automatically).

Jjepp Good Morning, if I may ask, I have a question regarding a worker prototype I've ...

Hello, I’m Allie!•3/15/24, 7:56 AM

It's not your responsibility to scale, it happens automagically. As for limits for WebSockets, your wallet is the limit.

jepp•3/15/24, 8:14 AM

@HardlyWorkin' let's say I go with the paid plan. It's $0.15/million requests per month. Could those 1 MIO websocket based requests be served also in an hour? Will the infrastructure scale?

Jjepp @HardlyWorkin' let's say I go with the paid plan. It's $0.15/million requests pe...

Hello, I’m Allie!•3/15/24, 8:16 AM

They could be served in a second, if needs be

Hello, I’m Allie!•3/15/24, 8:16 AM

If you are at that scale, you would probably have a custom pricing agreement, but

jepp•3/15/24, 8:17 AM

@HardlyWorkin' maybe... just scaling in theory. You never know. Realistically I am not expecting it. Just want to assess if the planned infrastructure is feasible

jepp•3/15/24, 8:19 AM

what does it mean in the logs if a bundled worker results into "canceled"? 50 ms cpu time exceeded and worker killed?

Jjepp what does it mean in the logs if a bundled worker results into "canceled"? 50 ms...

Hello, I’m Allie!•3/15/24, 8:20 AM

Cancelled iirc means the client has disconnected before a response was returned

jepp•3/15/24, 8:23 AM

hmm... i see the response from the websocket server in the clients console. So response was returned.

jepp•3/15/24, 8:26 AM

It might need further investigation

jepp•3/15/24, 8:26 AM

not a priority now

dave•3/15/24, 1:31 PM

What's that really popular open-source Workers project that provides a simple web UI for R2 uploading/sharing?

kian•3/15/24, 1:35 PM

https://r2explorer.dev/ ?

R2 Explorer

A Google Drive Interface for your Cloudflare R2 Buckets

dave•3/15/24, 1:36 PM

that's it, thank you!

ɢᴏᴅ||₳₦ł•3/15/24, 2:04 PM

i did wrangler login and it took me to a webpage and asked to allow consent,clicked on it and it said localhost refused to connect

ɢᴏᴅ||₳₦ł•3/15/24, 2:04 PM

how do i solve tht?

ɢᴏᴅ||₳₦ł•3/15/24, 2:07 PM

i am very new to this so sorry if i am in wrong channel

ɢᴏᴅ||₳₦ł•3/15/24, 2:27 PM

guys help pls

ɢɢᴏᴅ||₳₦ł i did wrangler login and it took me to a webpage and asked to allow consent,clic...

dave•3/15/24, 2:32 PM

you using Chrome by any chance?

Ddave you using Chrome by any chance?

ɢᴏᴅ||₳₦ł•3/15/24, 2:33 PM

yes ?

dave•3/15/24, 2:34 PM

I haven't spent any time on it, but I believe it's a new Chrome security measure. https://www.bleepingcomputer.com/news/google/new-google-chrome-feature-blocks-attacks-against-home-networks/

dave•3/15/24, 2:34 PM

try another browser

ɢᴏᴅ||₳₦ł•3/15/24, 2:34 PM

it will be tough

ɢᴏᴅ||₳₦ł•3/15/24, 2:34 PM

let me try

ɢᴏᴅ||₳₦ł•3/15/24, 2:35 PM

anything i can do in chrome only?

ɢᴏᴅ||₳₦ł•3/15/24, 2:37 PM

i tried on edge and it did not work

ɢɢᴏᴅ||₳₦ł i tried on edge and it did not work

dave•3/15/24, 2:44 PM

Edge is Chromium

ɢᴏᴅ||₳₦ł•3/15/24, 2:49 PM

i dont really have another browser

ɢᴏᴅ||₳₦ł•3/15/24, 3:33 PM

someone there?

ɢᴏᴅ||₳₦ł•3/15/24, 3:33 PM

/home/runner/Mini-Cricket/.config/npm/node_global/lib/node_modules/wrangler/wrangler-dist/cli.js:29573
            throw a;
            ^

Error: listen EADDRINUSE: address already in use 127.0.0.1:8976
    at Server.setupListenHandle [as _listen2] (node:net:1485:16)
    at listenInCluster (node:net:1533:12)
    at GetAddrInfoReqWrap.doListen [as callback] (node:net:1682:7)
    at GetAddrInfoReqWrap.onlookup [as oncomplete] (node:dns:109:8)
Emitted 'error' event on Server instance at:
    at emitErrorNT (node:net:1512:8)
    at process.processTicksAndRejections (node:internal/process/task_queues:82:21) {
  code: 'EADDRINUSE',
  errno: -98,
  syscall: 'listen',
  address: '127.0.0.1',
  port: 8976
}

/home/runner/Mini-Cricket/.config/npm/node_global/lib/node_modules/wrangler/wrangler-dist/cli.js:29573
            throw a;
            ^

Error: listen EADDRINUSE: address already in use 127.0.0.1:8976
    at Server.setupListenHandle [as _listen2] (node:net:1485:16)
    at listenInCluster (node:net:1533:12)
    at GetAddrInfoReqWrap.doListen [as callback] (node:net:1682:7)
    at GetAddrInfoReqWrap.onlookup [as oncomplete] (node:dns:109:8)
Emitted 'error' event on Server instance at:
    at emitErrorNT (node:net:1512:8)
    at process.processTicksAndRejections (node:internal/process/task_queues:82:21) {
  code: 'EADDRINUSE',
  errno: -98,
  syscall: 'listen',
  address: '127.0.0.1',
  port: 8976
}

getitng this error

rob•3/15/24, 3:39 PM

question- to use OpenAI streaming on a worker, do you need to use the openai NPM package?

You can't just use curl request and parse the. chunks for a SSE response to the frontend?

Can anyone elaborate for me.

Rrob question- to use OpenAI streaming on a worker, do you *need* to use the openai ...

dave•3/15/24, 4:36 PM

you don't have curl in a Worker. Use the openai npm package, it's much easier even if you aren't doing streaming.

rob•3/15/24, 4:43 PM

not curl, i meant fetch

rob•3/15/24, 4:43 PM

which is how im doing it now with a fetch to the cloudlfare gateway

rob•3/15/24, 4:43 PM

its woerking fne for me now

rob•3/15/24, 4:43 PM

just wanted to try streaming

Isaac McFadyen•3/15/24, 10:46 PM

You can manually parse the SSE, yes.

Isaac McFadyen•3/15/24, 10:46 PM

You'll just need to write the parsing logic yourself.

rob•3/15/24, 10:58 PM

Alright cool. Thank you

shaderkyle_ | SolSentryBot•3/15/24, 11:22 PM

hey, i noticed there is no support for ciphers in node:crypto when node_compat is enabled. Is there a polyfill available, and/or an equivalent implementation on subtle crypto, specifically with aes-256 and iv (initialization vectors)?

/home/runner/Mini-Cricket/.config/npm/node_global/lib/node_modules/wrangler/wrangler-dist/cli.js:29573 throw a; ^ Error: listen EADDRINUSE: address already in use 127.0.0.1:8976 at Server.setupListenHandle [as _listen2] (node:net:1485:16) at listenInCluster (node:net:1533:12) at GetAddrInfoReqWrap.doListen [as callback] (node:net:1682:7) at GetAddrInfoReqWrap.onlookup [as oncomplete] (node:dns:109:8) Emitted 'error' event on Server instance at: at emitErrorNT (node:net:1512:8) at process.processTicksAndRejections (node:internal/process/task_queues:82:21) { code: 'EADDRINUSE', errno: -98, syscall: 'listen', address: '127.0.0.1', port: 8976 }

Is there any sort of "on exit" or "on close" event callback for workers?

Similar Threads

Similar Threads

Similar Threads