I thought Cloudflare Builds had no coldstart at all, but I read that's 10s now?
I thought Cloudflare Builds had no coldstart at all, but I read that's 10s now?
L
LKinda in-line with GitHub Actions runners hosted on Azure to be fair.
CCloudflare workers don't (well.. mostly), maybe that's what you're thinking of?
Cthey advert Workers having no cold start in quite a few places but never builds as far as I know lol
Wyeah workers != builds
Wbuilds is it's own product but part of the workers suite
MDid you manage to fix it ?
MGreat tip, 100% agree with it
Jlol, i literally created a library yesterday for that reason
Jfor easy ssr in workers using html templates
Jnpm
A library for rendering dynamic HTML templates with embedded server-side JavaScript without using eval.. Latest version: 1.0.0, last published: 20 hours ago. Start using forge-templates in your project by running
npm i forge-templatesso Builds is just normal VMs, not V8 isolate
Wyeah, it's just a build server - it's basically just GitHub Actions and Workers is the app GitHub Actions deploys
SDoes anyone know if
"not_found_handling": "single-page-application"
KDoes Pages/Workers support using
pnpm
Syes you can
Khow does one do that?
S^ and make sure to use pnpm in your build script. either in the dashboard in wrangler
Kah no, I meant when building the site at cloudflare
Syes
Kah ic
Kis there a way to force it to use --frozen-lockfile or is it auto-set to do that?
Sas you see i dont defined that arg and it still used it
Sregarding this i was missing a catchall with
env.ASSETS.fetch"not_found_handling": "single-page-application"
AWhen using the new automated worker builds, how is it possible to access configured build secrets? For example I would like to execute drizzle migrations on push to main, before deploying to workers. But I don‘t get the database url within drizzle config currently!
HCloudflare Docs
Understand the different settings associated with your build.
SHi all, I have a question about Logpush and Workers work. I wanted to customize the sending of my Workers logs to Grafana, for this I found an option to set up a special Worker that will take requests from Logpush, format them and send them to Grafana Loki. I have enabled Worker Logs - Enabled and Logpush - Enabled in my Worker Logs, and I have a business subscription and a workers subscription in my account. I have set up a Job in Logpush, but I can only see test messages in Grafana (apparently they are created when the Job is created), but after that I don't see any messages. What could be the reason for this behavior? I also added console.log(); but it didn't help, I think I did everything according to the instructions but no positive result, has anyone experienced this?
AExactly!
HJust to confirm, did this answer your question?
AAh okay, sry I thought that was just to confirm which build configuration I meant. Question still there. Within the link you provided there‘s an section about build configuration and how to provide build time secrets and variables. I added a secret and try to access it via process.env.DATABASE_URL but it seems like there is no issue available. So I‘m wondering how to get access to this issue from within the drizzle.config.ts file to execute the Migration?
DFor a worker that is just an entry-point for other workers to invoke via RPC how come we have to add a fetch method?
How come we cant do something like:
How come we cant do something like:
SI'm following the code examples from the alarms API, where I only
setAlarm if getAlarm() returns nullgetAlarm() was consistently returning a date several weeks in the past though. Is this supposed to be possible? Does this mean that alarm reached its retry / backoff limit? What's the best practice in this case? Current workaround is to re-set the alarm if its date is too far in the past.
.I’m making a cf bot to handle http interactions that are sent by discord (a bot, or app if you want). When discord sends an interaction, I return a new response. Is it possible to see the response discord sends back about the response I sent to discord? (If that makes sense)
.I’m dumb
TWill make a "help" topic if need be but I have some headers being set in _headers file which I can see in the CF dashboard. Looks as expected, but these headers are not being set when viewing with ZAP test or $ curl -I <domain>.
Any idea why CF might be picking themup from the file ok but then not applying to the actual deployment?
Any idea why CF might be picking themup from the file ok but then not applying to the actual deployment?
TPerhaps it just takes some time to propegate?
JBrowser Rendering (cloned puppeteer) used to have a weird color shift when using 
this.client.on("Page.screencastFrame". I was procrastinating on creating a issue on GitHub but it looks like it's been silently fixed! Just wanted to say many many thanks on that one to who ever was 
RHow to create Astro Template Minimal ?
npm create cloudflare@latest -- mycode --framework=astro
npm create cloudflare@latest -- mycode --framework=astro
EInteresting RPC behavior that doesn't seem to be documented anywhere... The caller gets terminated with "Exceeded CPU Time Limits" error after 15 mins wall time (~30ms CPU time) and the callee (server) shows "Client Disconnected" with ~40ms CPU time in the metrics. The callee produces healthy logs for 15 mins, waking up every 30 seconds to check a job status and to extend its life with setTimeout. The caller works fine if the job takes less than 15 mins wall time or no RPC is used.
SHi everyone,
I'm new to this Discord, so I’m not sure if this is the right place to ask.
I’ve built a TypeScript worker that scrapes GPU prices hourly from multiple retailers using cheerio and stores the data for historical tracking. However, it's making too many subrequests—navigating from an index page to each product page is hitting Cloudflare CPU limits.
I tried using queues but I'm unsure how to implement them effectively. Does anyone have advice on managing this more efficiently?
I'm new to this Discord, so I’m not sure if this is the right place to ask.
I’ve built a TypeScript worker that scrapes GPU prices hourly from multiple retailers using cheerio and stores the data for historical tracking. However, it's making too many subrequests—navigating from an index page to each product page is hitting Cloudflare CPU limits.
I tried using queues but I'm unsure how to implement them effectively. Does anyone have advice on managing this more efficiently?
AQueues will work, but I’d personally use Workflows (in beta, as a note): https://developers.cloudflare.com/workflows/
- you can presumably have a workflow per retailer or product or w/e and just kick off the Workflows from a Worker Cron or something
Cloudflare Docs
Workflows is a durable execution engine built on Cloudflare Workers. Workflows allow you to build multi-step applications that can automatically retry, persist state and run for minutes, hours, days, or weeks. Workflows introduces a programming model that makes it easier to build reliable, long-running tasks, observe as they progress, and progra...
SThanks, i will look into this!
MHello, Im working on a minimal astro landing page, however I need to index a second page which needs to install
npm install react-datepicker , if my main/landing page is already deployed to cloudflare pages, how might install react-datapicker? is it through command line? any advise is welcome...
Eis there anyway to have a custom install command on cf workers?
RWhen I start a POST request from my local machine to a Cloudflare network, Cloudflare will not send the POST to my backend until it receives the entire POST body. Is there a way to change this behavior? I would like to have my local machine stream to my backend.
SPlease do not post your question in multiple channels/post it multiple times per the rules at #
welcome-and-rules. It creates confusion for people trying to help you and doesn't get your issue or question solved any faster.
welcome-and-rules. It creates confusion for people trying to help you and doesn't get your issue or question solved any faster.