hi am new to clould flare and learning, i have deployed my first default worker and as soon as i dep

hi am new to clould flare and learning, i have deployed my first default worker and as soon as i deployed it it already had like 20 requests, without me doing anything, is that normal ? what are those requests ?

Aabuff hi am new to clould flare and learning, i have deployed my first default worker ...

Cyb3r-Jak3•5/19/24, 1:58 PM

Bots scrape the SSL certificate logs and send requests to new hosts.

CCyb3r-Jak3 Bots scrape the SSL certificate logs and send requests to new hosts.

abuffOP•5/19/24, 2:00 PM

Thanks, they are sure fast starting scraping, like instant, hope they do not eat up too much of the free tier

Aabuff Thanks, they are sure fast starting scraping, like instant, hope they do not eat...

Cyb3r-Jak3•5/19/24, 2:02 PM

It can help to disable the workers.dev subdomain for a bit when first creating the worker. Means the bots will get rejected when they first try it

neoney•5/19/24, 3:52 PM

for storing profile pictures and other kinds of pictures (like product images uploaded), should I use R2 or Cloudflare Images (stored)?

neoney•5/19/24, 3:52 PM

which is more worth it?

CCyb3r-Jak3 It can help to disable the workers.dev subdomain for a bit when first creating t...

abuffOP•5/19/24, 5:05 PM

Thanksss for the tip !

skoogeer•5/19/24, 9:12 PM

Hey everyone, MailChannels will be discontinuing our free API service as of June 30th. Details are in this knowledgebase article: https://support.mailchannels.com/hc/en-us/articles/26814255454093-End-of-Life-Notice-Cloudflare-Workers

I would post this in the community forum, but for some reason access was blocked.

Help Center

End of Life Notice - Cloudflare Workers

Since launching our free email sending API for Cloudflare Workers customers, we have been proud to provide a simple yet effective solution to deliver messages from Workers code with a minimum of se...

skoogeer•5/19/24, 9:13 PM

Starting next week, you may notice a small fraction (1%) of your connections are rejected with an error message containing a link to the same announcement. Any messages that are accepted by the service will be delivered as usual and we will continue to provide support as usual until the final day of service.

Sskoogeer Hey everyone, MailChannels will be discontinuing our free API service as of June...

James•5/19/24, 9:40 PM

Ah, good to hear. We've been advising people to avoid this partnership for quite some time, so it's good to get an official End of Life notice.

Sskoogeer Hey everyone, MailChannels will be discontinuing our free API service as of June...

James•5/19/24, 10:06 PM

Not sure why you can't access the community, but I've made the announcement there so folks are aware: https://community.cloudflare.com/t/mailchannels-end-of-life-notice-migrate-now/658638

skoogeer•5/19/24, 10:07 PM

Thanks, @James. I figured you would take up the challenge.

James•5/19/24, 10:12 PM

Happy to help

JoeDoe•5/20/24, 5:00 AM

What's the best way to cache data from D1 used by a api request via workers?
I need simple data from the D1 database like userCount via a api request, but quering the database on each request doesn't seem optimal.
Is there a way to save the data in some global variable or some caching mechanism so the d1 database isn't queried often?

peterz•5/20/24, 9:21 AM

Hello, is it ok to use the same worker for scheduledscheduled and

fetch

fetch

? Are there any downsides to it?

zegevlier•5/20/24, 9:43 AM

That is perfectly fine to do. There aren't any real downsides, other than that the two calling methods will both be listed in analytics for the same worker. If your scheduled takes a lot more CPU time, that would throw off your averages for request CPU time usage.

avogadro•5/20/24, 1:09 PM

I noticed new error recently when deploying with upload_source_maps=trueupload_source_maps=true.

✘ [ERROR] A request to the Cloudflare API (/accounts/.../workers/scripts/...) failed.

  sourcemap index.js.map is too large, must not exceed 5MB [code: 10021]
  
  If you think this is a bug, please open an issue at: https://github.com/cloudflare/workers-sdk/issues/new/choose

✘ [ERROR] A request to the Cloudflare API (/accounts/.../workers/scripts/...) failed.

  sourcemap index.js.map is too large, must not exceed 5MB [code: 10021]
  
  If you think this is a bug, please open an issue at: https://github.com/cloudflare/workers-sdk/issues/new/choose

My worker is relatively small with a short list of dependencies (prisma, @planetscale/database, zod), and source map comes out to be around ~7MB. The 5MB limit seems very low. I had to disable source_maps for this worker, but I'd imaging a simple worker that is just proxying calls to a database is going to be a frequent use case of workers.

kaneki•5/20/24, 1:50 PM

ive made a normal worker with eventemitter and it works fine as long as its in the same file, when i put the bot.on()bot.on() function in an separate file it stops to work why is that?

kaneki•5/20/24, 1:51 PM

the bot.on()bot.on() function in a separate file, index.jsindex.js and castam.jscastam.js are in the same directory

kaneki•5/20/24, 2:38 PM

as far as i can see eventemitter doesn't work between 2 files...

Isaac McFadyen•5/20/24, 3:11 PM

It looks to be because castam.jscastam.js imports from index.jsindex.js but Workers always run the index.jsindex.js, so castamcastam never gets imported itself.

Isaac McFadyen•5/20/24, 3:12 PM

For it to work the import would need to go the other way around somehow (index.js importing castam.js)

IIsaac McFadyen For it to work the import would need to go the other way around somehow (index.j...

kaneki•5/20/24, 3:25 PM

Ohhh i seee

IIsaac McFadyen For it to work the import would need to go the other way around somehow (index.j...

kaneki•5/20/24, 3:31 PM

okay so now i made a emitter.jsemitter.js which has Eventemitter and ive imported the emitter.js into index.js and castam.js, still its like this...

kaneki•5/20/24, 3:31 PM

doesnt work i mean

kaneki•5/20/24, 3:32 PM

ignore the typo in bot.emit() its "hello" there ive fixed it

rob•5/20/24, 4:11 PM

is it best to do a service bindings or the new native rpc calls for worker to worker

Walshy•5/20/24, 4:21 PM

You'd need to use RPC over service bindings, RPC just provide an interface rather than the worker to worker part

rob•5/20/24, 5:38 PM

yah i re-read the blog from april, makes sense now

zensin•5/20/24, 7:45 PM

Need more RPC and stream r/w examples.FEels like I'm deciphering ancient spells from the few surviving pages of necronomicon

Luke•5/20/24, 9:02 PM

I'm getting a lot of Worker exceeded CPU time limit Worker exceeded CPU time limit errors - is it safe to change from bundledbundled to StandardStandard? Is there anything that changes other than longer CPU limits?

Luke•5/20/24, 10:09 PM

Also, are there any tips to help figure out what code the CPU limits are caused by?

LLuke I'm getting a lot of ` Worker exceeded CPU time limit` errors - is it safe to ch...

kian•5/20/24, 10:17 PM

Limits only go up

kian•5/20/24, 10:17 PM

Subrequests go from 50 to 1,000, CPU time goes from 50ms to 30s, I don't think there's really anything else

LLuke Also, are there any tips to help figure out what code the CPU limits are caused ...

kian•5/20/24, 10:18 PM

There isn't really any good solution for it - you can try timing with Date.now()Date.now() but that is wall time so if you're doing subrequests to websites or bindings then you'd see a long time but the CPU isn't really being used just waiting for a response.

Kkian There isn't really any good solution for it - you can try timing with `Date.now(...

Luke•5/20/24, 10:21 PM

Okay I see, thank you. I recall that there was an issue with Date.now()Date.now() not progressing outside of

async

async

methods. Is that still a problem do you know?

kian•5/20/24, 10:21 PM

I meant for local development, Date.now() isn't limited there

kian•5/20/24, 10:22 PM

In deployed Workers it will still only progress on IO

Luke•5/20/24, 10:23 PM

Ah ok that makes sense. Thanks for the help

kian•5/20/24, 10:24 PM

Most Workers only really run into CPU issues when parsing a lot of data or stuff like WASM

Luke•5/20/24, 10:26 PM

I'm thinking I might be hitting CPU limits with the HTMLRewrtierHTMLRewrtier but having a hard time proving it. Its tough to know when this issue started happening too since the exception causes the worker to crash and not send any logs.

Luke•5/20/24, 10:27 PM

I was thinking we could put another worker infront of this worker to detect if there are any issues. Do worker subrequests share the same memory/cpu limits?

Kurzon•5/20/24, 10:31 PM

How can I prevent Cloudflare Workers from counting Unauthorized accesses as requests?

Kurzon•5/20/24, 10:31 PM

I am limited to 100,000 requests per day.

Kurzon•5/20/24, 10:32 PM

But it's not fair because someone can easily attack the website and send 100k requests.

Kurzon•5/20/24, 10:32 PM

So hence why I have an API Key so the website can only be authorized/accessed with it.

Kurzon•5/20/24, 10:32 PM

But it seems that Cloudflare Workers counts unauthorized accesses as requests, so someone can easily spam requests to my Workers site.

Kurzon•5/20/24, 10:32 PM

And hit 100k with spams, and boom, I can't use it anymore.

KKurzon So hence why I have an API Key so the website can only be authorized/accessed wi...

kian•5/20/24, 10:54 PM

Is the API key static or do you look them up in a database somewhere else?

Kkian Is the API key static or do you look them up in a database somewhere else?

Kurzon•5/20/24, 11:08 PM

API key is static

hi am new to clould flare and learning, i have deployed my first default worker and as soon as i dep

Similar Threads

Similar Threads

Similar Threads