This is what I have concluded, we need to update the `expiration` when we update the value, if you n

AAditya Mathur This is what I have concluded, we need to update the `expiration` when we update...

I

Ian•7/2/24, 6:21 AM

Sure, but when you

get('key')

get('key')

it returns an object with the

expiration

expiration

field, right? So if it already has that field, you can re-write it back without altering it.

I

Ian•7/2/24, 6:22 AM

When you're altering it, you're taking the

expiration

expiration

which is fixed in time, and keep adding to it each time the request happens. is that what you want? for the expiration to be moved forward?

A

Aditya MathurOP•7/2/24, 6:22 AM

If you update the expiration on each request, you will then be putting values less than 60 seconds which will not be getting registered

A

Aditya MathurOP•7/2/24, 6:22 AM

All I am concluding from this is WorkerKV cannot be used for rate limiting in general because the API is not build for this purpose. It is more for having global config for env or feature flags

I

Ian•7/2/24, 6:22 AM

it could be; I have a lateral solution...

A

Aditya MathurOP•7/2/24, 6:23 AM

I am listening

I

Ian•7/2/24, 6:23 AM

make a record for each request with a ttl of 15 minutes.

I

Ian•7/2/24, 6:23 AM

then list and count all the records on each request to see if they exceed the limit.

I

Ian•7/2/24, 6:24 AM

the maximum number of records you'll have is the

limit

limit

. so not very many.

I

Ian•7/2/24, 6:24 AM

and you never have to write to them after the first

put()

put()

; you fire and foget them until the next request where you

list()

list()

and count.

I

Ian•7/2/24, 6:25 AM

basically use the namespace as a "leaky bucket"

IIan then list and count all the records on each request to see if they exceed the li...

A

Aditya MathurOP•7/2/24, 6:26 AM

Yes you can manually work this out i.e. without the use of

expiration

expiration

prop. But at the end of the day it will be too expensive to use this solution for rate limiting.

IIan and you never have to write to them after the first `put()`; you fire and foget ...

A

Aditya MathurOP•7/2/24, 6:27 AM

Because the limit is 1,000/day. Check this https://developers.cloudflare.com/kv/platform/pricing/

Cloudflare Docs

Pricing · Cloudflare Workers KV

Workers KV is included in both the Free and Paid Workers plans.

I

Ian•7/2/24, 6:29 AM

You could use D1 https://developers.cloudflare.com/d1/platform/pricing/

Cloudflare Docs

Pricing · Cloudflare D1 docs

D1 bills based on:

I

Ian•7/2/24, 6:29 AM

There's also durable objects.

IIan There's also durable objects.

A

Aditya MathurOP•7/2/24, 6:29 AM

Yes, that is what I am thinking about to drop the support for

WorkerKV

WorkerKV

and use

DurableObjects

DurableObjects

AAditya Mathur Yes, that is what I am thinking about to drop the support for `WorkerKV` and use...

I

Ian•7/2/24, 6:31 AM

the kv plan for paid customers still makes sense.

AAditya Mathur Yes, that is what I am thinking about to drop the support for `WorkerKV` and use...

I

Ian•7/2/24, 6:33 AM

https://developers.cloudflare.com/durable-objects/examples/build-a-rate-limiter/ here's an actual example of building a rate limiter with durable objects

IIan https://developers.cloudflare.com/durable-objects/examples/build-a-rate-limiter/...

A

Aditya MathurOP•7/2/24, 6:34 AM

Nice, will be building something like this then

I

Ian•7/2/24, 6:35 AM

https://community.cloudflare.com/t/creating-a-rate-limiter-using-workers-kv/319256 there is also this forum post about using WorkerKV for rate limiting

I

Ian•7/2/24, 6:36 AM

KV is not really a good idea here due to it’s eventually consistent nature. You can however make a rate-limit from Durable Objects 272, which some others have already made from too.

IIan > KV is not really a good idea here due to it’s eventually consistent nature. Yo...

A

Aditya MathurOP•7/2/24, 6:36 AM

Yeah

A

Aditya MathurOP•7/2/24, 6:37 AM

Now I know why is that

V

vyastj•7/2/24, 10:05 AM

Hello

Can someone help me here or point to correct channel ?

workers-ai

X

Xiangmian•7/3/24, 7:42 AM

I can't log in either

LLawrence Hi, wrangler login always got error, did anybody knows how to fix that? ``` Synt...

B

Bọc Link VN•7/3/24, 8:52 AM

I got the same today when use npx wrangler login

LLawrence Hi, wrangler login always got error, did anybody knows how to fix that? ``` Synt...

I

ivxenog•7/3/24, 9:17 AM

same here.

H

Hard@Work•7/3/24, 9:43 AM

@twodft @Xiangmian @crypto Howdy! The team is aware, and they are working on a fix

V

Vero•7/3/24, 2:15 PM

Hey. This should be fixed now, can you please confirm?

R

Ross•7/3/24, 4:29 PM

npx wrangler pages deploy

npx wrangler pages deploy

is erroring for me, Cloudflare having API issues?

 ✘ [ERROR] Received a malformed response from the API

  upstream connect error or disconnect/reset before headers. retried and the latest reset reason: conn... (length = 114)
  GET /accounts/***/pages/projects/example -> 503 Service Unavailable

 ✘ [ERROR] Received a malformed response from the API

  upstream connect error or disconnect/reset before headers. retried and the latest reset reason: conn... (length = 114)
  GET /accounts/***/pages/projects/example -> 503 Service Unavailable

 ✘ [ERROR] Received a malformed response from the API

  upstream connect error or disconnect/reset before headers. retried and the latest reset reason: conn... (length = 114)
  GET /accounts/***/pages/projects/example -> 503 Service Unavailable

 ✘ [ERROR] Received a malformed response from the API

  upstream connect error or disconnect/reset before headers. retried and the latest reset reason: conn... (length = 114)
  GET /accounts/***/pages/projects/example -> 503 Service Unavailable

RRoss `npx wrangler pages deploy` is erroring for me, Cloudflare having API issues? `...

C

Chaika•7/3/24, 4:35 PM

it is yea, I think somewhat calmed down now if you want to retry or wait a bit

A

aj•7/4/24, 3:35 PM

Hello! I'm having an issue when trying to specify my account id, even with it defined in my wrangler.toml, it still is returning an error.

✘ [ERROR] More than one account available but unable to select one in non-interactive mode.
Please set the appropriate `account_id` in your `wrangler.toml` file.
Available accounts are (`<name>`: `<account_id>`):

✘ [ERROR] More than one account available but unable to select one in non-interactive mode.
Please set the appropriate `account_id` in your `wrangler.toml` file.
Available accounts are (`<name>`: `<account_id>`):

✘ [ERROR] More than one account available but unable to select one in non-interactive mode.
Please set the appropriate `account_id` in your `wrangler.toml` file.
Available accounts are (`<name>`: `<account_id>`):

✘ [ERROR] More than one account available but unable to select one in non-interactive mode.
Please set the appropriate `account_id` in your `wrangler.toml` file.
Available accounts are (`<name>`: `<account_id>`):

W

Walshy•7/4/24, 3:35 PM

what's your wrangler.toml?

A

aj•7/4/24, 3:36 PM

D

Divyendu•7/4/24, 3:38 PM

Can I refer to env vars in wrangler.toml? I am setting up inngest with Cloudflare and need to set INNGEST_SIGNING_KEY, technically that's a "secret" but setting it via

wrangler secret put INNGEST_SIGNING_KEY

wrangler secret put INNGEST_SIGNING_KEY

doesn't make it available in the "environment variables", inngest sync fails and to put it in environment variables, I need to add it to wrangler.toml as "plain text" and deploy.

Aaj Click to see attachment

W

Walshy•7/4/24, 3:40 PM

everything in wrangler toml is safe to share fwiw but regardless, assuming that's at the top of the file - seems fine but i'd need the whole thing to say for sure

A

aj•7/4/24, 3:41 PM

That is the whole thing.. Thanks anyway :)

S

sadsciencee•7/5/24, 6:38 AM

is it possible to still use .dev.vars localy while using wrangler.toml to manages pages deployment? not seeing anything in the docs and when I run pages dev it just defaults to the 'preview' environment (which is more of a staging env for me, not meant for local dev)

A

Arpit•7/5/24, 7:24 AM

Hey all, I need help, I am not able to run any wrangler project, not even a fresh template, all of them giving this error

PS D:\others\durable-objects-template> wrangler dev

 ⛅️ wrangler 3.63.1
-------------------

Your worker has access to the following bindings:
- Durable Objects:
  - COUNTER: Counter
⎔ Starting local server...
╭─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ [b] open a browser, [d] open Devtools, [l] turn off local mode, [c] clear console, [x] to exit
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
C:\Users\user\AppData\Roaming\npm\node_modules\wrangler\wrangler-dist\cli.js:29765
            throw a;
            ^

MiniflareCoreError [ERR_RUNTIME_FAILURE]: The Workers runtime failed to start. There is likely additional logging output above.
    at #assembleAndUpdateConfig (C:\Users\user\AppData\Roaming\npm\node_modules\wrangler\node_modules\miniflare\dist\src\index.js:9178:13)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
    at async Mutex.runWith (C:\Users\user\AppData\Roaming\npm\node_modules\wrangler\node_modules\miniflare\dist\src\index.js:3521:16) {
  code: 'ERR_RUNTIME_FAILURE',
  cause: undefined
}

Node.js v20.14.0

PS D:\others\durable-objects-template> wrangler dev

 ⛅️ wrangler 3.63.1
-------------------

Your worker has access to the following bindings:
- Durable Objects:
  - COUNTER: Counter
⎔ Starting local server...
╭─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ [b] open a browser, [d] open Devtools, [l] turn off local mode, [c] clear console, [x] to exit
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
C:\Users\user\AppData\Roaming\npm\node_modules\wrangler\wrangler-dist\cli.js:29765
            throw a;
            ^

MiniflareCoreError [ERR_RUNTIME_FAILURE]: The Workers runtime failed to start. There is likely additional logging output above.
    at #assembleAndUpdateConfig (C:\Users\user\AppData\Roaming\npm\node_modules\wrangler\node_modules\miniflare\dist\src\index.js:9178:13)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
    at async Mutex.runWith (C:\Users\user\AppData\Roaming\npm\node_modules\wrangler\node_modules\miniflare\dist\src\index.js:3521:16) {
  code: 'ERR_RUNTIME_FAILURE',
  cause: undefined
}

Node.js v20.14.0

PS D:\others\durable-objects-template> wrangler dev

 ⛅️ wrangler 3.63.1
-------------------

Your worker has access to the following bindings:
- Durable Objects:
  - COUNTER: Counter
⎔ Starting local server...
╭─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ [b] open a browser, [d] open Devtools, [l] turn off local mode, [c] clear console, [x] to exit
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
C:\Users\user\AppData\Roaming\npm\node_modules\wrangler\wrangler-dist\cli.js:29765
            throw a;
            ^

MiniflareCoreError [ERR_RUNTIME_FAILURE]: The Workers runtime failed to start. There is likely additional logging output above.
    at #assembleAndUpdateConfig (C:\Users\user\AppData\Roaming\npm\node_modules\wrangler\node_modules\miniflare\dist\src\index.js:9178:13)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
    at async Mutex.runWith (C:\Users\user\AppData\Roaming\npm\node_modules\wrangler\node_modules\miniflare\dist\src\index.js:3521:16) {
  code: 'ERR_RUNTIME_FAILURE',
  cause: undefined
}

Node.js v20.14.0

PS D:\others\durable-objects-template> wrangler dev

 ⛅️ wrangler 3.63.1
-------------------

Your worker has access to the following bindings:
- Durable Objects:
  - COUNTER: Counter
⎔ Starting local server...
╭─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ [b] open a browser, [d] open Devtools, [l] turn off local mode, [c] clear console, [x] to exit
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
C:\Users\user\AppData\Roaming\npm\node_modules\wrangler\wrangler-dist\cli.js:29765
            throw a;
            ^

MiniflareCoreError [ERR_RUNTIME_FAILURE]: The Workers runtime failed to start. There is likely additional logging output above.
    at #assembleAndUpdateConfig (C:\Users\user\AppData\Roaming\npm\node_modules\wrangler\node_modules\miniflare\dist\src\index.js:9178:13)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
    at async Mutex.runWith (C:\Users\user\AppData\Roaming\npm\node_modules\wrangler\node_modules\miniflare\dist\src\index.js:3521:16) {
  code: 'ERR_RUNTIME_FAILURE',
  cause: undefined
}

Node.js v20.14.0

A

Arpit•7/5/24, 7:25 AM

what am i missing ... ?

R

Royal•7/5/24, 8:57 AM

Hello, I have a question about running

wrangler dev --remote

wrangler dev --remote

is it possible to only specify to run R2 remotely but other service to be run locally?

Original message was deleted

R

Royal•7/5/24, 9:26 AM

Thank you for the clarifications

J

João Castro•7/7/24, 2:22 AM

I'm trying to use a service binding from a Cloudflare Pages project to a Worker, and the binding does not work locally if the Pages dev server is started before the Worker. From what i read in the docs, this doesn't seem to be the intended behavior

M

MadManick•7/7/24, 8:29 AM

Hello, I have a question with environments and wrangler.

I need to configure a d1 database for preview and production environments separately, but want to use the same binding name for both. That way the code would be the same but connected database would be different.

Right now, I am trying something like this

[[d1_databases]]
binding = "DB"
database_name = "name"
database_id = "<uuid>"
migrations_dir = "./migrations"

[env.preview]
[[d1_databases]]
binding = "DB"
database_name = "name-preview"
database_id = "<uuid>"
migrations_dir = "./migrations"

[[d1_databases]]
binding = "DB"
database_name = "name"
database_id = "<uuid>"
migrations_dir = "./migrations"

[env.preview]
[[d1_databases]]
binding = "DB"
database_name = "name-preview"
database_id = "<uuid>"
migrations_dir = "./migrations"

[[d1_databases]]
binding = "DB"
database_name = "name"
database_id = "<uuid>"
migrations_dir = "./migrations"

[env.preview]
[[d1_databases]]
binding = "DB"
database_name = "name-preview"
database_id = "<uuid>"
migrations_dir = "./migrations"

[[d1_databases]]
binding = "DB"
database_name = "name"
database_id = "<uuid>"
migrations_dir = "./migrations"

[env.preview]
[[d1_databases]]
binding = "DB"
database_name = "name-preview"
database_id = "<uuid>"
migrations_dir = "./migrations"

But this errors out while deploying with "Failed to publish your Function. Got error: the binding name 'DB' cannot be used more than once". Is there a way to do this?

Original message was deleted

M

MadManick•7/7/24, 8:58 AM

Thanks! That worked!

Original message was deleted

J

João Castro•7/7/24, 8:00 PM

Then how are circular dependencies handled? Its not possible to run two workers that have Service Bindings to each other locally?

J

João Castro•7/7/24, 8:04 PM

Sounds good! I'll give it a go soon

This is what I have concluded, we need to update the `expiration` when we update the value, if you n

Similar Threads