if you have one Cron trigger every minute, will a new cron be triggered even before the previous one
if you have one Cron trigger every minute, will a new cron be triggered even before the previous one is done?
DI hear the coolest people in the world are in that program!
PYeah that would be my main motivation for participating honestly (the town-halls and such), but I don't think I'll be applying bc I'm just not the target audience
ANot sure to be honest, it would be easy to test by setting one up and just having it sleep for over a minute
Xhi all. I have a question about using WebRTC from a worker. Is it possible to make an outbound UDP connection? Is there a WebRTC API?
BJust for the relevant team's information, while the 520 error issue is now resolved (for the last couple of hours), it seems the worker is no longer actually respecting the smart placement. I have a D1 database binding in
WNAM
WYeah correct, we have currently mitigated by disabling placement for non-GETs -- we're working on making it fully functional again
BAh ok, thanks.
DHey guys. I use workers to host my website. I have several thousand files which are connected to R2. I recently got charged a lot since the website was having a million hits a day. I do have caching enabled on the domain but I don’t think it’s working. You guys know of any solution for this?
JThere are too much issues of "ERR_SSL_VERSION_OR_CIPHER_MISMATCH" on newly created workers, even on raw url <name>.abc.workers.dev
.Hi, can someone please help me with installing sentry... I've been trying for days with no luck. I made a project, I can get it to work with picking nodejs on their dashboard, but the integrations on the workers part just won't work, it links with this post: https://community.cloudflare.com/t/workers-integrations-unlisted-sentry-projects/693813
AJust my 2c, but I would recommend https://baselime.io/ instead - it was bought by Cloudflare and Sentry is genuinely terrible in my experience and Baselime will work in a few clicks
Baselime
Use logs, metrics, distributed tracing and wide-events to detect, diagnose and resolve issues before they impact your users.
WWe're back to full smart placement operation @Brett Willis
S@Ashley How does baselime play once installed if you're using terraform?
AI'm not sure to be honest, not tried it 
RCloudflare Turkiye 
856 kb 1 minute o.O
856 kb 1 minute o.O
RThis usually happens in the evenings, especially on weekends.
RWhen we ask Cloudflare, they say it is from the Turkish ISP. When we ask the Turkish government, they say it is from Cloudflare.
RDo you think the IP address that Cloudflare gives to a free site and Discord is the same?
HTry it out:
1DNS over Discord: A records
discord.com A @9.9.9.9 +noall +answer +shortdiggy diggy hole
•9/2/24, 6:32 PM
RI have observed that in all free sites, even when the cloudflare page & workers are online a lot, the sites load very slowly, especially the images in them are much slower. But of course this is not the case in discord, in that case discord does not use cloudflare.
HIt may just be bad routing. Cloudflare does prioritize certain datacenters for paid customers
RMy friend, as you can see in the photo, it is a simple site, a logo.png and a background image, what's wrong with that? Shouldn't we add an image to our site?
RHey, since you can send emails from workers, how i can add DKIM to them and how to sign them?
LI'm building a pages web app, and currently have two workers, one for notifications for the web app (SSE) and another for emails, they all use the same D1 database, does this seem reasonable? or how would you simplify?
AImo this is very much an architectural choice. There’s no real “right” and “wrong”. If these two workers belong to the same “system” (as in a serverless environnment, you’ll often have multiple components that make up a system - in your case, maybe it’s two workers) then it makes sense. If they change together also, a good reason for them to live together. However, if they are only sharing the DB for convenience and these two workers are parts of different systems (or should be) then it would probably be more architecturally-sound if they were more decoupled (e.g. web app writes a message to a queue that the email consumer picks up, will all info needed in the payload). My 2c anyway, without fully knowing your app etc.
MTwo questions. Is there a way to use Browser Rendering in Workers but from an specific country? Most websites are different depending on the user location. And also, is there a pricing page for Browser Rendering? Thanks
.How does time work in workers if the request is handled “closest” to the user? What would Date.now() return?
.Because the machine will be in a certain country with a certain time zone
.Oh .now is utc nvm
WAll time in Cloudflare is UTC
XHey everyone! I'm looking into building a medium-complexity app fully on CF workers/dev platform. This will be 1st time for me. The app will be SPA + workers backend (maybe also a couple specific server-rendered pages, not sure yet) using also D1, R2 and queues. My backend background is in frameworks like ASP.NET Core (C#) and Cake/Symfony (PHP).
Q: What are the frameworks that are recommended to use in this case? I've stumbled upon hono, but it definitely feels lackluster compared to what I'm used to. I understand serverless and/or edge requires/encourages lightweight-ness, but maybe there is some middle ground? I do not need max performance possible.
Q: What are the frameworks that are recommended to use in this case? I've stumbled upon hono, but it definitely feels lackluster compared to what I'm used to. I understand serverless and/or edge requires/encourages lightweight-ness, but maybe there is some middle ground? I do not need max performance possible.
SI was working with Cloudflare Workers + bun & Hono and was using Knex as my orm. I was successfully ably to make a connection on Cloudlfare Pages to Knex+Postgres with Connection Pooling however on Workers I am recieving a TypeErorr.
I tried using both the connectionString and specifiying the individual parameters for the database. Additionally I set node_compat = true
and also added the prefix previously when using compatibility_flags = ["nodejs_compat"]
I also played around with the version numbers of both the sql connection adapter and
Error:
Error Source:
I tried using both the connectionString and specifiying the individual parameters for the database. Additionally I set node_compat = true
and also added the prefix previously when using compatibility_flags = ["nodejs_compat"]
I also played around with the version numbers of both the sql connection adapter and
Error:
Error Source:
Salso as a side question is hyperdrive required for node-postgres queries?
Iis having api key in my workers code a bad idea?
HCloudflare Docs
Store sensitive information, like API keys and auth tokens, in your Worker.
Iworkers code is private, right? so whats the need?
HWell, for one, anyone with access to your account can just read the API Key
HFor Secrets, once they are set, they cannot be viewed from the dashboard directly
Ianyone with access to your accountthats just me, right? (i also have 2fa)
HAssuming that you don't invite anyone else to manage parts of your account, nor share an API key for your CF account with anyone else
.Why does a response from a worker sometimes show not secure?
JWhat do you mean? Are you accessing the worker over http, or https?
.Through a route on my domain, yes
JHttp, or https? If you access over http, most browsers show a ‘not secure’ warning nowadays
.Now it isn’t showing not secure, but it was earlier
.https
JThat’s unlikely. If you can reproduce with a https link, please let us know
SHi, i have a question about Workers. If I redirect a request to another domain(my domain but in a different zone) I always get CF_CONNECTING_IP - 2a06:98c0:3600::103 in the header, as I understand this is normal CF behavior. Is there any way I can “bypass” this rule so that I can find out the real IP address of the person making the request?
