I think in theory this should just work always, if retries are unlimited and get their own cpu time,

CChaika I think in theory this should just work always, if retries are unlimited and get...

D

Diogo Ferreira•10/25/24, 4:07 PM

You should be alright!

S

Sam•10/25/24, 4:30 PM

Currently I’m using queues for some ai functions that have multiple steps.

The reason I use queues as it’s a mobile app and I don’t want the user to wait too long to get an initial response that the work has started.

I write to my Postgres jobs table to say the job has started (ideally I would replace this with the workflow instance to avoid an unneeded network call)-> Processes a photo of a meal with a vision model -> queries my nutrition db -> generates macros nutrients and calories -> once succeeds or fails updated my job in my Postgres db

My frontend client subscribes via a web socket to the status of the job

Would I be better just moving this to workflows (and getting rid of the queue) as it seems to have queueing built in?

Latency is pretty important in my use case as these models take awhile and the user is expecting it to be processed under 10 seconds.

(So would be great essentially if I could reduce the need to add those additional round trips to update the job in my db by subscribing to the workflow instance instead)

A

ajgeiss0702•10/25/24, 5:40 PM

(for billing) what counts as an invocation? does each step attempt count as an invocation, or is it only one invocation for the whole workflow instance?
(please ping me if you reply)

Anyone searching for this later: this was kindof partially answered here: workflows

AAhmad Awais also 500 limit on workflow is too small. Workflows for platforms?

A

Advany•10/25/24, 6:09 PM

That would be really awesome!!!

F

fry69•10/25/24, 6:11 PM

Not yet though -> would workflows work for workers for

A

apolodoro•10/25/24, 6:47 PM

Can a wait step have a webhook that can trigger to stop the wait and continue?
An example usecase would be an alarm system which should only be trigger if a machine is stopped for more than an hour. The workflow would begin when the machine stops and then if it turns on again before that hour an external function would call the workflow webhook and cancel the workflow (or even better cause it to take different actions than the default path).

M

makisuo•10/25/24, 8:52 PM

I keep getting this error locally with Workflows any idea?

An RPC stub was not disposed properly. You must call dispose() on all stubs in order to let the other side know that you are no longer using them. You cannot rely on the garbage collector for this because it may take arbitrarily long before actually collecting unreachable objects. As a shortcut, calling dispose() on the result of an RPC call disposes all stubs within it.

An RPC stub was not disposed properly. You must call dispose() on all stubs in order to let the other side know that you are no longer using them. You cannot rely on the garbage collector for this because it may take arbitrarily long before actually collecting unreachable objects. As a shortcut, calling dispose() on the result of an RPC call disposes all stubs within it.

An RPC stub was not disposed properly. You must call dispose() on all stubs in order to let the other side know that you are no longer using them. You cannot rely on the garbage collector for this because it may take arbitrarily long before actually collecting unreachable objects. As a shortcut, calling dispose() on the result of an RPC call disposes all stubs within it.

An RPC stub was not disposed properly. You must call dispose() on all stubs in order to let the other side know that you are no longer using them. You cannot rely on the garbage collector for this because it may take arbitrarily long before actually collecting unreachable objects. As a shortcut, calling dispose() on the result of an RPC call disposes all stubs within it.

Original message was deleted

M

makisuo•10/25/24, 9:26 PM

code is a bit convoluted will try to get a small reproduction

C

Chief•10/26/24, 12:39 AM

Do retries count as a separate invocation for billing?

CChief Do retries count as a separate invocation for billing?

D

Dmitry•10/26/24, 1:30 AM

I would imagine so

D

Dmitry•10/26/24, 1:31 AM

oh invocation... hmm good question

D

Dmitry•10/26/24, 1:33 AM

I hope concurrency controls are on the way! So we can say like.. only x workflows (per workflow type) can run concurrently

D

Dmitry•10/26/24, 1:35 AM

For example I think Resend limits API requests to like 40req/sec I think? So limiting the concurrency of a workflow that sends emails would be great

D

Dmitry•10/26/24, 1:36 AM

Would be cool if you could set concurrency at a step level too.. not just the overall workflow

S

sdr•10/26/24, 3:58 AM

The api docs for instance creation has a bug in the response shape. Result is a single object and not a collection like the docs suggest
https://developers.cloudflare.com/api/operations/wor-create-new-workflow-instance

Cloudflare API Documentation

Interact with Cloudflare's products and services via the Cloudflare API

T

tornado•10/26/24, 5:11 AM

Is it possible to have workflows defined in a pages worker?

Ttornado Is it possible to have workflows defined in a pages worker?

D

DanGamble•10/26/24, 7:24 AM

No, they're just workers at the moment. This is just my opinion but they likely will stay that way with the introduction of worker assets (https://developers.cloudflare.com/workers/static-assets/)

Cloudflare Docs

Static Assets | Cloudflare Workers docs

Create full-stack applications deployed to Cloudflare Workers.

D

DanGamble•10/26/24, 9:12 AM

With a nested step, if a nested step fails or needs to be retried, does that mean the outer step also fails/retries?

D

DanGamble•10/26/24, 9:13 AM

Also, love the "Rules of Workflows" piece of the documentation https://developers.cloudflare.com/workflows/build/rules-of-workflows/. Definitely helps understand things a lot more. Would love to see this kind of thing on all products. Especially DO's

Cloudflare Docs

Rules of Workflows | Cloudflare Workflows docs

A Workflow contains one or more steps. Each step is a self-contained, individually retriable component of a Workflow. Steps may emit (optional) state that allows a Workflow to persist and continue from that step, even if a Workflow fails due to a network or infrastructure issue.

L

LordSilver•10/26/24, 11:13 AM

Workflows not available in the TF provider?

C

Cayter•10/26/24, 4:41 PM

is there no way to delete workflow...?

CCayter is there no way to delete workflow...?

F

fry69•10/26/24, 4:43 PM

Not currently it seems -> workflows

Ffry69 Not currently it seems -> https://discord.com/channels/595317990191398933/129692...

C

Cayter•10/26/24, 4:44 PM

oh thx...

Aajgeiss0702 (for billing) what counts as an invocation? does each step attempt count as an i...

A

ajgeiss0702•10/26/24, 6:57 PM

bump

M

makisuo•10/26/24, 10:03 PM

Workflows work fine for me in dev mode but deployed seem to never finish, eventhough the last expected step finished succesfully

M

makisuo•10/26/24, 10:09 PM

I am using EffectTS and for that reason am wrapping step.do to use it better with effect but dont really see how that could be the issue:

export const stepEffect = <A, E>(
    step: { do: (name: string, fn: () => Promise<any>) => Promise<any> },
    stepName: string,
    effectFn: Effect.Effect<A, E, never>,
): Effect.Effect<A, E, never> => {
    return Effect.promise(() => executeRpcWithCleanup(() => step.do(stepName, async () => Effect.runPromise(effectFn))))
}

export const stepEffect = <A, E>(
    step: { do: (name: string, fn: () => Promise<any>) => Promise<any> },
    stepName: string,
    effectFn: Effect.Effect<A, E, never>,
): Effect.Effect<A, E, never> => {
    return Effect.promise(() => executeRpcWithCleanup(() => step.do(stepName, async () => Effect.runPromise(effectFn))))
}

export const stepEffect = <A, E>(
    step: { do: (name: string, fn: () => Promise<any>) => Promise<any> },
    stepName: string,
    effectFn: Effect.Effect<A, E, never>,
): Effect.Effect<A, E, never> => {
    return Effect.promise(() => executeRpcWithCleanup(() => step.do(stepName, async () => Effect.runPromise(effectFn))))
}

export const stepEffect = <A, E>(
    step: { do: (name: string, fn: () => Promise<any>) => Promise<any> },
    stepName: string,
    effectFn: Effect.Effect<A, E, never>,
): Effect.Effect<A, E, never> => {
    return Effect.promise(() => executeRpcWithCleanup(() => step.do(stepName, async () => Effect.runPromise(effectFn))))
}

Mmakisuo I am using EffectTS and for that reason am wrapping step.do to use it better wit...

I

Isaac McFadyen•10/26/24, 10:41 PM

Your first step for troubleshooting should probably be to see if it works without EffectTS. I don't think anyone here uses EffectTS and so the amount of help we can give when that's involved is going to be limited

IIsaac McFadyen Your first step for troubleshooting should probably be to see if it works withou...

M

makisuo•10/26/24, 10:44 PM

yeah will remove some and troubleshoot a bit myself

Just wanted to see if someone had run into similar issues before.

A

Avi•10/27/24, 12:06 AM

how do workflow updates work?

A

Avi•10/27/24, 12:07 AM

like if i deploy a new version of the code

A

Avi•10/27/24, 2:13 AM

ah, i found a thread addressing my concerns here: workflows

D

Dmitry•10/27/24, 2:45 AM

Anyone try implementing workflows into a SvelteKit app yet?

Trying to figure out how to load a Workflow class in something other than the workflow-starter app (i.e -

src/index.ts

src/index.ts

src/index.ts

src/index.ts file)

B

Beaconcha.in•10/27/24, 3:16 AM

Hi

B

Beaconcha.in•10/27/24, 3:19 AM

Hi

A

Avi•10/27/24, 5:16 AM

is it possible to define different workflows for different environments, like
env.Production
env.Production
env.Production
env.Production vs
env.Preview
env.Preview
env.Preview
env.Preview in
wrangler.toml
wrangler.toml
? it is!

A

Avi•10/27/24, 6:21 AM

lol

LLordSilver I guess that's a Cloudflare thing then, because your Worker would still be activ...

C

CanRau•10/27/24, 11:24 AM

Think they explain how it works here https://blog.cloudflare.com/workers-pricing-scale-to-zero

CCanRau Think they explain how it works here <https://blog.cloudflare.com/workers-pricin...

L

LordSilver•10/27/24, 2:34 PM

Yeah I remember this, but that's Workers, not Workflows

L

LordSilver•10/27/24, 2:34 PM

unless a Workflow runs on a Worker?

LLordSilver unless a Workflow runs on a Worker?

I

Isaac McFadyen•10/27/24, 2:40 PM

It does, yes.

I

Isaac McFadyen•10/27/24, 2:40 PM

In the Workers runtime anyway.

DDmitry Anyone try implementing workflows into a SvelteKit app yet? 👀 Trying to figure...

M

Matt Silverlock•10/27/24, 3:20 PM

Bind to a Workflow from your SvelteKit app and invoke it? The Workflow doesn’t have to be in the same script.

S

sdr•10/27/24, 11:05 PM

This error message is confusing to me, throwing a NonRetryableError means the error is handled by the CF workflows platform right?

The execution of the Workflow instance was terminated, as a step threw an NonRetryableError and it was not handled

The execution of the Workflow instance was terminated, as a step threw an NonRetryableError and it was not handled

The execution of the Workflow instance was terminated, as a step threw an NonRetryableError and it was not handled

The execution of the Workflow instance was terminated, as a step threw an NonRetryableError and it was not handled

S

sdr•10/27/24, 11:05 PM

So it is handled but its by CF and not me

Ssdr So it is handled but its by CF and not me

I

itsmatteomanf•10/27/24, 11:11 PM

I'd add a "[...] and it was not handled by user's Worklow code, so the Workflow failed.", that should explain it better.

I think in theory this should just work always, if retries are unlimited and get their own cpu time,

Similar Threads

Similar Threads

Similar Threads