For how long after a workflow is finished, `WORKFLOW.get(id)` will return the finished instance? Als

For how long after a workflow is finished,

WORKFLOW.get(id)

WORKFLOW.get(id)

will return the finished instance? Also, does

.restart()

.restart()

work for finished workflows too?

Mm.muiznieks Yes, I've noticed the same thing for me as well. Also the response that request...

m.muiznieks•1/29/25, 7:46 PM

I am on the Workers paid plan.

For at least 3 - 4 days I see that the workflows are flowing... not as expected.

They build up a massive queue (which seems to grow bigger day by day) and have long inactivity periods (as visible on the screen—before that, there was no activity for 1h+). Meanwhile, usually I've not seen more than 28 instances running, in rare cases they hit at one moment higher, but havent seen near 100.

The workflow itself contains just one simple request (to Meta conversion API), so there is nothing resource-demanding (because I'm testing if and how to scale the workflow usage).

At the error logs I see that the requests might fail due to "Durable Object restarting," because of some code updates. Thats cool, but I'm just waiting when this queue will be "errored" when the wrong timing hits.

So my concerns and question is - are there some other limits that must be taken in mind when building the Workflows? Or is this some issue related to my account or what?

I understand that it's low possibility to get an answer here, but I still hope on that. :))

Thank you .

Mm.muiznieks I am on the Workers paid plan. For at least 3 - 4 days I see that the workflow...

m.muiznieks•1/29/25, 7:58 PM

Addition to previous post - here is the situation on another Flow - queue just getting bigger, but nothing is processed. It feels like that only one workflow connected to the worker is running. Did I miss this limit somewhere if its true?

Mm.muiznieks I am on the Workers paid plan. For at least 3 - 4 days I see that the workflow...

Seekerdasbatatas•1/29/25, 9:07 PM

Are there any other workflows in your account? Note that the concurrency limit is account wide

SSeekerdasbatatas Are there any other workflows in your account? Note that the concurrency limit i...

m.muiznieks•1/29/25, 9:41 PM

No. These are two working workflows, which basically do the same thing.
By looking into possible solutions I divided the requests in half.

That helped a bit, as the active one started to process something and now the queue is slowly decreasing, but the other one - does not do anything, and now is up to 2k in the queue (both waiting queue sums are the same as before and keeps increasing).

For wider context - they are receiving the requests from the Zaraz custom-managed component. But I don't know if it has anything to do with this unless that affects some other underlying processes.

Mm.muiznieks No. These are two working workflows, which basically do the same thing. By look...

Seekerdasbatatas•1/29/25, 9:42 PM

Huh weird, thanks for letting us know - can you send your account id for us to investigate?

SSeekerdasbatatas Huh weird, thanks for letting us know - can you send your account id for us to i...

m.muiznieks•1/29/25, 9:46 PM

Yes. As already sent before, here it is.

My account ID: 5e3dea7b807381cba36b9292bde14920

Seekerdasbatatas•1/29/25, 9:46 PM

Thanks, I'll look into it tomorrow morning

m.muiznieks•1/29/25, 9:49 PM

I guess this issue is similar as what's already posted here: workflows @Rakhim

Jonathan Riche•1/30/25, 12:06 AM

Do Cloudflare workers bindings/apis work inside Workflows ie the AI binding?

Original message was deleted

Jonathan Riche•1/30/25, 1:00 AM

Awesome thx

SSeekerdasbatatas Thanks, I'll look into it tomorrow morning

m.muiznieks•1/30/25, 1:41 AM

What I can add now after looking at the tendency, it is somehow related to the incoming requests to the workflow. Now, after the peak hours are over, the processing tempo has increased proportionally to the decrease in new requests.

I hope this somehow helps investigate further and find a solution.

DD Trombett For how long after a workflow is finished, `WORKFLOW.get(id)` will return the fi...

D TrombettOP•1/30/25, 11:26 AM

Bumping this in case anyone knows

DD Trombett Empty message

Seekerdasbatatas•1/30/25, 11:45 AM

On limits, we advertise 3 days on free and 30 days on paid (although this is not enforced atm, but could change at any moment)(https://developers.cloudflare.com/workflows/reference/limits/)

.restart()

.restart()

is not implemented yet

Cloudflare Docs

Limits · Cloudflare Workflows docs

Limits that apply to authoring, deploying, and running Workflows are detailed below.

D TrombettOP•1/30/25, 11:46 AM

Ok, thanks!

kchro3•1/30/25, 9:48 PM

Hi folks, we're seeing that if you execute a sleep step, it takes a few minutes to change states from running to waiting, and that backs up the queue pretty significantly. Would someone be able to explain in more detail why this would happen under the hood and what we should expect?

Mitya•1/31/25, 1:12 PM

Hi - I'm new to Workflows. I'm a little lost in the various ways of doing things and am trying to understand how Workflows differs from other approaches to task distribution. In what situations would Workflows be preferable instead of, say, worker service bindings, or queues? Both allow tasks to be distributed and run asynchronously. A really simple use-case example would be appreciated.

Mm.muiznieks What I can add now after looking at the tendency, it is somehow related to the i...

Seekerdasbatatas•1/31/25, 6:53 PM

To keep you posted, the bug has been identified and will be rolling out throughout next week probably

Luke•1/31/25, 7:27 PM

Is there an easy way to debug workflows in vs code? In my workers project I can just use the

javascript debug terminal

javascript debug terminal

and it works fine. For some reason it doesn't w/ workflows - all of my breakpoints are "unbound"

LLuke Is there an easy way to debug workflows in vs code? In my workers project I can ...

Luke•1/31/25, 7:35 PM

Disregard - i forgot that w/ workers I don't use the

javascript debug terminal

javascript debug terminal

I have to run

wrangler dev

wrangler dev

and then attach vs code to the process. The same works fine w/ workflows.

Luke•1/31/25, 7:36 PM

Next question I have is - how can I restart a failed workflow? I tried using

instance.restart

instance.restart

instance.terminate

instance.terminate

, etc. and it says

Not implemented yet

Not implemented yet

. Any tips on that?

Kkchro3 Hi folks, we're seeing that if you execute a sleep step, it takes a few minutes ...

Seekerdasbatatas•1/31/25, 7:58 PM

It's an implementation detail for workflows - we need to allow some time for top level work, this time might get reduced before GA, but for beta we were more generous because of correctness (because it's JS, you can do some really weird stuff with control flow in a relatively easy way

)

Seekerdasbatatas•1/31/25, 7:59 PM

We hopefully are

a bunch of stuff next week that will make the queuing aspect better :soontm:

Luke•2/1/25, 6:10 AM

I'm trying to use Browser Rendering in my workflow but debug the code because:

I can't run browser rendering locally (need to use
```
--remote
```
```
--remote
```
)
I can't run workflows using using
```
--remote
```
```
--remote
```

Any good work around for this? I'm trying to avoid having multiple projects like pushing to a queue to have have Browser Rendering pick it up and process it. It feels like a perfect synergy for Workflows to work with Browser Rendering but the development lifecycle might be too painful at the moment to need to debug via deploying the code and using

console.log

console.log

crunchy•2/1/25, 10:46 PM

By the way, workflows do not trigger for me locallly (dev) when created by worker. Works in prod. wrangler 3.107.2

crunchy•2/1/25, 10:47 PM

Also what sort of concurrency is likely to be allowed in future, are these things intended for use at scale?

Dæ•2/1/25, 11:44 PM

Can someone give me an intro to this… I’ll ask ChatGPT

Ashkan•2/1/25, 11:49 PM

guys how can I get access to env variable in workflow?

Ashkan•2/1/25, 11:50 PM

export class myclass extends WorkflowEntrypoint<Env, Params> {
    async run(event: WorkflowEvent<Params>, step: WorkflowStep) {
    
    }
}

export class myclass extends WorkflowEntrypoint<Env, Params> {
    async run(event: WorkflowEvent<Params>, step: WorkflowStep) {
    
    }
}

AAshkan `export class myclass extends WorkflowEntrypoint<Env, Params> { async run(ev...

Ashkan•2/1/25, 11:52 PM

I found it as a comment in example code:

// Can access bindings on this.env

// Can access bindings on this.env

Fernando Barroso•2/2/25, 1:27 AM

two questions:

1) Is there an architectural icon for cf workflows yet?
2) anybody know where can I get a library of icons that I could use to build diagrams like these:

FFernando Barroso two questions: 1) Is there an architectural icon for cf workflows yet? 2) anybo...

Isaac McFadyen•2/2/25, 1:48 AM

For 2) see https://cfdata.lol/icons/ (unofficial)

Isaac McFadyen•2/2/25, 1:48 AM

Not sure about 1.

Fernando Barroso•2/2/25, 2:24 AM

Oh that site is pretty sick - thank you!

apolodoro•2/2/25, 12:14 PM

Is there a way to limit concurrent workflows based on their id? For example if the id starts with 'type1' only one workflow can be running at the same time

Aapolodoro Is there a way to limit concurrent workflows based on their id? For example if t...

Matt Silverlock•2/2/25, 2:41 PM

Not yet. Looking at controls here by Workflow - e.g. a

max_concurrent_instances

max_concurrent_instances

config.

Ccrunchy Also what sort of concurrency is likely to be allowed in future, are these thing...

Matt Silverlock•2/2/25, 2:42 PM

Next big step is 4,500 concurrent (very soon) and we want to get to 10k (at least) over the next few weeks/this quarter.

Making sure it all works + nothing bottlenecks.

MMatt Silverlock Not yet. Looking at controls here by Workflow - e.g. a `max_concurrent_instances...

apolodoro•2/2/25, 2:56 PM

Nice but not yet implemented right?

Aapolodoro Nice but not yet implemented right?

Seekerdasbatatas•2/3/25, 11:53 AM

correct

SSeekerdasbatatas correct

Jadugar_Jaggu•2/3/25, 4:13 PM

Can I know the limit of memory per workflow ?

For workers it's 128 MB shared per invocation right?

JJadugar_Jaggu Can I know the limit of memory per workflow ? For workers it's 128 MB shared p...

Seekerdasbatatas•2/3/25, 4:15 PM

it's the same 128MiB on Workflows too (and depending on scheduling, it maybe be shared with other instances)

SSeekerdasbatatas it's the same 128MiB on Workflows too (and depending on scheduling, it maybe be ...

Jadugar_Jaggu•2/3/25, 4:16 PM

Ok it means every step will get 128 MiB or a whole workerflow ?

JJadugar_Jaggu Ok it means every step will get 128 MiB or a whole workerflow ?

Seekerdasbatatas•2/3/25, 4:17 PM

the whole workflow

SSeekerdasbatatas the whole workflow

Jadugar_Jaggu•2/3/25, 4:20 PM

Ok thanks, well this is somewhat concerning if there are some tasks which use memory too much.

That's why may be my workflow steps are throwing an error of cpu/memory limit after doing work of around 128 MiB ( nearly )

But on retry it's again working till 128 MiB

what's this behaviour?

D TrombettOP•2/3/25, 6:07 PM

Docs say

step.sleep

step.sleep

duration is in seconds if it's a number but it seems to me that it's actually ms?

D TrombettOP•2/3/25, 7:26 PM

ah I see, thanks

Victor•2/4/25, 9:44 AM

I'm unable to use cross-script calls to workflows with wrangler dev (local)

Getting

Worker "workflows:AfterContractActivated-development"'s binding "USER_WORKFLOW" refers to a service "core:user:delivery-development", but no such service is defined.

Worker "workflows:AfterContractActivated-development"'s binding "USER_WORKFLOW" refers to a service "core:user:delivery-development", but no such service is defined.

Is that not supported? It doesn't seem specified in the docs

laurynas•2/4/25, 1:31 PM

has there been any discussion on setting error callbacks for workflow steps? I’d love an

onError

onError

callback i could pass where i could set some custom logic to try when a step errors. E.g.:

export class TextGenerator extends WorkflowEntrypoint<CloudflareBindings> {
  async run(
    event: Readonly<WorkflowEvent<{ prompt: string }>>,
    step: WorkflowStep,
  ) {
    const { prompt } = event.payload;

    return await step.do(
      "generate-text",
      {
        retries: { limit: 3, delay: 1000, backoff: "exponential" },
        onError: async (error, retry) => {
          if (error instanceof AITimeoutError) {
            // Complex model timed out, try with a faster one first
            try {
              return await generateText({
                model: "gpt-3.5-turbo",
                prompt,
                timeout: 5000,
              });
            } catch {
              // If faster model fails, retry with original complex model (retry is the original callback in step.do)
              return retry();
            }
          }
          throw error;
        }
      },

export class TextGenerator extends WorkflowEntrypoint<CloudflareBindings> {
  async run(
    event: Readonly<WorkflowEvent<{ prompt: string }>>,
    step: WorkflowStep,
  ) {
    const { prompt } = event.payload;

    return await step.do(
      "generate-text",
      {
        retries: { limit: 3, delay: 1000, backoff: "exponential" },
        onError: async (error, retry) => {
          if (error instanceof AITimeoutError) {
            // Complex model timed out, try with a faster one first
            try {
              return await generateText({
                model: "gpt-3.5-turbo",
                prompt,
                timeout: 5000,
              });
            } catch {
              // If faster model fails, retry with original complex model (retry is the original callback in step.do)
              return retry();
            }
          }
          throw error;
        }
      },

Llaurynas has there been any discussion on setting error callbacks for workflow steps? I’d...

Matt Silverlock•2/4/25, 4:46 PM

Isn’t that just another step? Write an outer step that encapsulates both.

Any

onError

onError

callback would still have to be treated like another step, but ends up being another not-quite-a-step API with its own semantics.

Run stepA, catch the error, call optional stepB, etc.

export class TextGenerator extends WorkflowEntrypoint<CloudflareBindings> { async run( event: Readonly<WorkflowEvent<{ prompt: string }>>, step: WorkflowStep, ) { const { prompt } = event.payload; return await step.do( "generate-text", { retries: { limit: 3, delay: 1000, backoff: "exponential" }, onError: async (error, retry) => { if (error instanceof AITimeoutError) { // Complex model timed out, try with a faster one first try { return await generateText({ model: "gpt-3.5-turbo", prompt, timeout: 5000, }); } catch { // If faster model fails, retry with original complex model (retry is the original callback in step.do) return retry(); } } throw error; } },

For how long after a workflow is finished, `WORKFLOW.get(id)` will return the finished instance? Als

Similar Threads

Similar Threads

Similar Threads