I would remove the cache and remove the 2nd db connection that doesnt use hyperdrive and it should b
I would remove the cache and remove the 2nd db connection that doesnt use hyperdrive and it should be faster 
Uand if theres some reason you really dont want to use hyperdrive for a specific route I would make it where you return only the hyperdrive connection or only the direct connection instead of opening both
ZWhy remove the cache? It seems to be much faster. Just trying to understand
Uhyperdrive already adds a cache so that would just double cache now
ZTrue, but for some reason the Hyperdrive cache is way slower. It adds about 80ms of latency
ZWould cloudflare D1 be better for my use case? I just want the lowest latency possible
ID1 would be lower latency in the location where it's created, but it's also SQLite (not Postgres) so anything Postgres specific won't work with D1.
DI am using photon-rs using WebAssembly, and the worker throws
Error 1102: Workers Resources Limit
EHi, everyone. Sorry for bothering you, but I need your help
I hosted a static website using Cloudflare workers to optimize the performance
how can I increase the page speed? what should I do, like configuring or anything else?
I hosted a static website using Cloudflare workers to optimize the performance
how can I increase the page speed? what should I do, like configuring or anything else?
JI just switched to paid workers plan. Is there something like a safety limit or something, so I don't end up spending big money by a mistake, since it's paid on additional requests as-you-go?
CIf your site is purely static then I would recommend pages
UThere's no hard limits you can impose, setting up WAF rules to try and block any bot traffic and rate limiting rules to disallow spam would be things you can do to try and ensure expected usage levels.
Normally you could also setup notifications for when you reach a certain request amount but seems due to the billing changes recently made to workers you cannot do this yet with the updated billing model.
Normally you could also setup notifications for when you reach a certain request amount but seems due to the billing changes recently made to workers you cannot do this yet with the updated billing model.
CNuxtJS SSR worker, my code is fetching my dotnet endpoint in azure using a cloudflare origin certificate. I have strict TLS turned on and my nuxt worker is getting 522 errors on request to dotnet endpoint. Do i need to somehow get worker to do an origin pull? If i turn off strict TLS worker goes back to functional
Mhi I can't find the api to create a new worker https://developers.cloudflare.com/api/operations/worker-script-delete-worker ? how do i create a new worker from api. Thank
Interact with Cloudflare's products and services via the Cloudflare API
Emy website is the build of react. how to deploy it using pages
EReact is a popular framework for building reactive and powerful front-end applications, built by the open-source team at Facebook.
WWhat are your go to rest api framework for cf worker?
EPersonally I use Hono
WThanks! I’ll try to look at it
WStill a newbie with serverless concept
WDoes each api call invoke a function and each invocation spawn a worker, is it still preferable to write async?
WAnd how about service which have to wait for user input, let’s say an otp service.
- User hit api
- Api send otp via sms
- Api wait for 30 sec for user inputting otp
ZEach time a request is made to your worker, the fetch() function is ran with the request information. In workers, multiple requests can share the same isolate (the thing that actually runs the code). You should write workers using async because if you don't do that, you wont be able to do many operations, as they often require async.
I would re-frame the problem here as instead of having the server wait for the user to input the OTP, the server give the user a time limit of 30 seconds to enter the OTP. There are many ways you could do this. You could send the user a JWT that they need to send back that expires after 30 seconds, or you could store the login attempt in a datastore like #d1-beta or #durable-objects.
I would re-frame the problem here as instead of having the server wait for the user to input the OTP, the server give the user a time limit of 30 seconds to enter the OTP. There are many ways you could do this. You could send the user a JWT that they need to send back that expires after 30 seconds, or you could store the login attempt in a datastore like #d1-beta or #durable-objects.
WThanks! Now it’s more clear.
I just see durable object, it looks like redis cache
which I think suit the otp use case.
But the jwt method looks more interesting, how does it work? Isn’t jwt token is too long for otp which is usually only four to six digit?
I just see durable object, it looks like redis cache
which I think suit the otp use case.But the jwt method looks more interesting, how does it work? Isn’t jwt token is too long for otp which is usually only four to six digit?
EI believe whats being suggested is the user is presented with a jwt and then within 30 seconds they must return both the jwt and the totp code, which mitigates the need to maintain any state yourself
WWow! That’s genius
WI wish I can be as smart as both of you @zegevlier @Erisa | Support Engineer
WMany thanks for the help 
ZI don't think that matters here? Why would you need secrecy for this?
Zoh like that, yea you shouldn't store the OTP in there. Good call, a JTW wouldn't work well here
ZPoint still stands though, you just need something that encrypts the information stored
WThis is so much to digest, but I think I get a glimpse of the concern 
so basically, the user can give back the the jwt token without having to recieve the otp?
so basically, the user can give back the the jwt token without having to recieve the otp?
Sany thoughts on what can cause
#<Response> could not be cloned.? I have error logging with Sentry but I'm pretty sure the stack trace is not accurate -.-SSentry says it is coming from this line
await this.state.storage.put(foo, bar);Snvm we had a bug that was trying to serialize a request into DO storage, so Sentry was right this time
FHey everyone! Are there any plans to allow query parameters via route patterns without also applying the wildcard to the path? For example:
example.com/page* is the only way to support https://example.com/page?blah=1, but it also matches https://example.com/page1234. This seems like a pretty serious limitation.
Sit's not possible to match the ?
FIt returns an error saying query strings aren’t allowed in the pattern.
CThere's been nothing mentioned about them changing soon and Worker Routes haven't changed in a while, that's also a documented limitation:
https://developers.cloudflare.com/workers/configuration/routing/routes/#matching-behavior
It's not the most optimal solution but you could parse the URL in the Worker and check if the pathname equals the one you're looking for, otherwise just passthrough
https://developers.cloudflare.com/workers/configuration/routing/routes/#matching-behavior
Route pattern matching considers the entire request URL, including the query parameter string. Since route patterns may not contain query parameters, the only way to have a route pattern match URLs with query parameters is to terminate it with a wildcard, *.So I doubt it would change anytime too soon.
It's not the most optimal solution but you could parse the URL in the Worker and check if the pathname equals the one you're looking for, otherwise just passthrough
return fetch(request) the request, wouldn't interfere with anything but would cause Worker Invocations stillRoutes allow users to map a URL pattern to a Worker. When a request comes in to the Cloudflare network that matches the specified URL pattern, your …
CDepending on how your website is setup, you could also potentially take advantage of the more specific matching behavior. If it's something like you want /post?ggg
/post* to go to a worker but not /posts?ggg, if you made a route for /posts* service none, that should be more specific and stop the Worker from running on that pathFThanks - just wanted to see if anybody has heard anything about a change. Appreciate the reply
Cis anyone here using Nuxt/NitroServer with Service Bindings on their worker?
Mquestion about the workers.. styandard requests
Mi upgraded to premium, is it unlimited now?
UIf by "premium" you mean workers paid, then yes you can do an unlimited number of requests.
10 million requests included + $0.30/million after
30 million cpu milliseconds included + $0.02/million after
10 million requests included + $0.30/million after
30 million cpu milliseconds included + $0.02/million after
Mcant i just higher my subscription to not have additional pay
UI mean enterprise plans give custom pricing but its not going to be you pay $x for fully unlimited requests and cpu usage afaik
USo no you cant pay more for a flat rate there will always be usage based costs
Mwell 10million is not much
