is calling a worker from another worker not possible? I'm using fetch to call worker A from worker B
is calling a worker from another worker not possible? I'm using fetch to call worker A from worker B
UYou should use service bindings
UService bindings are an API that facilitate Worker-to-Worker communication via explicit bindings defined in your configuration.
Aawesome, thank you!
If worker B is a web framework (
I bind it as a service in worker A (
If worker B is a web framework (
Honohttps://worker_b.me.workers.dev/${param}I bind it as a service in worker A (
env.B), how do I call it and pass paramworked when I passed the whole worker URL 
env.B.fetch(https://worker_b.me.workers.dev/${param})Athank you very much @Unsmart | Tech debt
HHey guys. We have been looking at workers for a while and have been interested in migrating our backend to workers. I have a couple concerns though which I needed some guidance with. For starters, our backend right now is a complex express app with many routes, middleware, controllers etc. I have looked at Hono and have liked what they have to offer, so im sure we'll be fine in terms of refactoring our app to use Hono instead of express. My main concern though is that we use libraries such as @sendgrid/mail, @aws-sdk/client-s3, mongoose, speakeasy, sharp, multer-s3, firebase-admin, just to name a few. I'm aware that workers doesnt support a full node.js environment yet. Do you guys think it will be possible to migrate to workers?
HIt definitely won’t be a one-click migration(and some of the libraries may have to be rewritten from scratch), but I don’t see anything that would prevent it at the moment
Syeah to be on the safe side I used a fake path on our real domain
DIt’s 100% working for me with the invalid domain. Could you try using a custom cache key?
DI’ll check again to see if I’m overlooking something.
DI’m using with a period.
DAh I see I already shared this, my bad.
DThat sorta makes sense, wouldn’t be surprised if there’s some very light URL validation being done before caching a URL.
D@lekensteyn have you looked into how purging the Cache API globally (for a single URL) works with a fake domain name?
DIs the Cache API shared among all Workers in a zone?
Eyes
Dooo thank you, was about to hit myself in the face with that one
Dthoughts on itty vs hono? hono claims to be 2x faster
UMost people here use hono afaik but both are good and will probably be exactly the same performance wise on workers 
Dthanks
UPretty sure it will throw if it fails so yes
SThanks a lot! What is the relationship between miniflare and Wrangler by the way? Being new to Cloudflare Workers this has been somewhat confusing to me.
HHey all. What would be the best way to initialize connections to external services, i.e in my example to MongoDB, when using Hono with workers? Right now, the only way ive been able to do it (using the module worker mode) is by creating custom middleware that initializes a realm app, logs in, and connects to the DB before every request that needs to interact with the DB. Idk how practical this approach is though. Any ideas?
HIdeally, I would like to connect to the database once and then listen for requests
SCool! Thanks a lot for the clarification!:)
D@lekensteyn thank you both for your advice, here's what I got working to purge the Worker cache both locally and globally for a URL. 
https://gist.github.com/Manouchehri/8db5ebcecbbc6b286ef226cb57122b1d
I'm using
https://gist.github.com/Manouchehri/8db5ebcecbbc6b286ef226cb57122b1d
I'm using
https://ai.moda.invalid/hash_here as my URL, I can confirm it's working (only when using a custom domain though afaik, not .workers.dev as expected).DI'm not entirely sure if the
cache.delete(cacheKey) is actually needed, or if purge_cache will always have cleared the local cache just as fast.EIt wouldnt be immediate, the global cache purge can take a number of seconds to go through
DI have a worker that reverse-proxies a blog on a subdirectory. It works perfect on the worker preview url, but it redirects to the origin blog instead of proxying it when I use my custom domain.
I am very sure I am just missing a little detail on how workers work.
I am very sure I am just missing a little detail on how workers work.
DI have the worker code and detailed explanation in the #workers-help (currently the topmost question) if that's needed
ATo deploy a worker using Terraform, is the only module that works https://registry.terraform.io/providers/cloudflare/cloudflare/latest/docs/resources/worker_script?
I'm asking as script seems a bit... misleading / odd, as modern workers often have several files, not just one "script". Anything else in TF modules doesn't look suitable though, so am I missing something, or is it just not possible with TF and we have to use Wrangler, for example?
I'm asking as script seems a bit... misleading / odd, as modern workers often have several files, not just one "script". Anything else in TF modules doesn't look suitable though, so am I missing something, or is it just not possible with TF and we have to use Wrangler, for example?
NHey I am not able to "create service" button like this
NWhenever I click on mange worker I got this page plz help me 
EThe dashboard has been changed around since that tutorial was made, click "Create Worker" on the page you end up on and you should be on the right track
E@NaveenSingh900
NIt creates service and creates worker is same?
EYes
RWhen I log into my account and click worker routes, the manager workers button is disabled. Is that because I don’t have permission to manage or create workers?
I’ve installed the wrangler CLI locally and tried to deploy a worker, which led to a 10000 auth error. But I show as having write permission with wrangler whoami. Am I missing something?
I’ve installed the wrangler CLI locally and tried to deploy a worker, which led to a 10000 auth error. But I show as having write permission with wrangler whoami. Am I missing something?
Dis there a wrangler command to delete/reset the local kv database? can't find anything in the docs
