Can someone explain the difference between workers for platforms and standard? I couldnt unterstand
Can someone explain the difference between workers for platforms and standard? I couldnt unterstand it pretty well on internet. Thanks
AI have a worker that runs every minute, checks if there are posts newer than the last post saved in KV, and forwards them if so. On the past few days, I've noticed posts being forwarded twice, which indicates that KV is returning outdated info a minute later. I've never seen this happen before; was there a recent change that would make KV take longer to update? Is there anything I can do to avoid this happening, other than changing the interval the worker runs on? (please ping with response)
WWorkers for platforms is so you can turn workers into a SaaS product.
Docs give a good explanation: https://developers.cloudflare.com/cloudflare-for-platforms/workers-for-platforms/
Docs give a good explanation: https://developers.cloudflare.com/cloudflare-for-platforms/workers-for-platforms/
ECan i use them as normal workers? I just need the cf properties and better limits
WWonder if this is what we're seeing in #releases
I haven't had time yet to investigate.
There shouldn't have been any changes which could cause this but of you can confirm that behaviour I'd recommend posting in #kv
I haven't had time yet to investigate.
There shouldn't have been any changes which could cause this but of you can confirm that behaviour I'd recommend posting in #kv
WWfP doesn't give you more limits (except more script count) and the cf properties is available in all workers
WWhat are you hitting exactly?
EWell i have my own basic dashboard for cloudflare API and manage few hundreds of domains. Each of them should have its own worker and each worker should save headers etc from each visitor. Would this be possible with normal workers? - also there is a limit of 500 workers per account right? Can i have more workers with WFP
WI'd recommend doing that in a single Worker but err yeah, you can have unlimited with WfP so if you really wanted to go that route it'd work
EWhat would you recommend for saving the information of the visitors. Should i use the hypersrive, d1… or maybe something else
WDepends on the information
WIf it's for analytics and you just want to save like request count - Analytics Engine
WIf you need an actual DB (e.g. auth) then yeah D1 or your own DB with Hyperdrive
EOk thank you very much i appreciate it.
Ayeah looks the same - I think I first noticed it on 5/20. should I just repost that same message in #kv?
KIs an event handler (like fetch()) is always required even if you only use that worker in a service binding via RPC? I can't seem to deploy without specifying fetch.
Kyou don't need a fetch handler, but you do need a default export
Kif you're not exporting your WorkerEntrypoint class as the default, Wrangler will get upset
THello there!
I built a multiplayer game server with node.js that communicates with the clients via websockets and I was wondering whether cloudflare workers could be used to host it (currently I'm using a traditional vps):exitedCat:
I built a multiplayer game server with node.js that communicates with the clients via websockets and I was wondering whether cloudflare workers could be used to host it (currently I'm using a traditional vps):exitedCat:
DHey, workers (and more specifically, Durable Objects with their websocket hibernation api) would be perfect, assuming you can adapt your data storage solution. #durable-objects is the best channel to ask if you get stuck, the docs don't contain enough examples to understand everything
Tthanks for the quick answer :) I'll make same research regarding durable objects then
Jhow can i clear everything i have cached using the workers cache api cache.put()?? i tried a curl {purge_everything:true} using the api but the cached response is not being deleted
KI'm looking for an example of a successfully built workers-sdk / workerd stack --- I'm fixing a bug in workerd myself and would like to compile it end-to-end. Anybody have any examples / resources?
PQuestion regarding durable objects (DO). If I have one DO class but let’s say 100 instances. Will I be billed 100 x 128mb or only once 128mb if they all do in practice not occupy more than that? It is not explained very well in the documentation
Cloudflare Docs
Durable Objects are only available on the Workers Paid plan.
CThose footnotes go over it pretty well:
4 Duration is billed in wall-clock time as long as the Object is active, but is shared across all requests active on an Object at once. Once your Object finishes responding to all requests, it will stop incurring duration charges. Calling accept() on a WebSocket in an Object will incur duration charges for the entire time the WebSocket is connected. If you prefer, use state.acceptWebSocket() instead, which will stop incurring duration charges once all event handlers finish running.
5 Duration billing charges for the 128 MB of memory your Durable Object is allocated, regardless of actual usage. If your account creates many instances of a single Durable Object class, Durable Objects may run in the same isolate on the same physical machine and share the 128 MB of memory. These Durable Objects are still billed as if they are allocated a full 128 MB of memory.
Cif they were all constantly active, you would get billed 100 x 128 mb. Otherwise it's just a static billing for 128 mb x the time that instance is active
PSo I could read it as if they are still billed individually?
CWhat do you mean by
if they all do in practice not occupy more than that?
PLike. If each instance uses 24/7 1mb. Do I get billed 100 x 128mb?
CCorrect, as the footnote says
Duration billing charges for the 128 MB of memory your Durable Object is allocated, regardless of actual usage
PJust to confirm. The footnote means each DO instance and not DO class?
Cyes, I mean each instance of a DO class is, itself, a single Durable Object
PBecause, since they didn’t use the word each or individually in the end, I think the description is ambiguous.
At least for me it doesn’t seem wrong to assume, that they mean that all instances combined will always take at least 128 mb of memory
At least for me it doesn’t seem wrong to assume, that they mean that all instances combined will always take at least 128 mb of memory
CDurable Object Namespace is the name of the binding/the overall object
Ni have an cloudflare worker that fetches another api, can i send more than 100mb of size in that fetch call or does the upload limit of 100mb (cloudflare standarts) also count for worker sending files to stuff?
Nand if yes can i increase that?
KKI really want to use global scope for asynchoronous operations
KBut I cannot
KHow would you guys deal with this?
CStore that data in KV and read it on request
KBut i want it globally available, not just in the fetch function itself
CIf the JSON is local, then depending on the size of the JSON you can bundle it into the worker
https://developers.cloudflare.com/workers/wrangler/bundling/
https://developers.cloudflare.com/workers/wrangler/bundling/
AAnyone know of any npm libraries that work on CloudFlare Workers to edit images? I'm looking to specifically change colors (convert image to duo-tone). I looked through the CF Images product but that doesn't seem to allow the editing I need. A library like
sharp would be great, but that doesn't seem to work as far as I can tell.An/m, found a great one: @cf-wasm/photon
Ehi anyone know "zaraz context enricher worker"? I created one, but it didn't work at all
Vhello guys we are trying to run Workers on a small Shopify Website, is it possibile to active the O2O feature without purchase an enterprise license?
VIs there something alternative that we could do in order to run Workers/proxy on a shopify website?
SHello everyone, I'm trying to create a discord bot using workers. but for some odd reason, when i try to save the endpoints url in discord developer portal, it throws me with an Invalid Form Body, and doesn't let me change the interactions endpoint.
.Do you handle the PING type and the verify the request
