I've recently built a solution on Workflows, and I'm trying to identity if I'm hitting some sort of

AAustin B I've recently built a solution on Workflows, and I'm trying to identity if I'm h...

M

If you’re firing off 31 requests that each trigger hundreds of Workflows - you’re likely hitting rate limits: https://developers.cloudflare.com/workflows/reference/limits/

Cloudflare Docs

Limits

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

M

Matt Silverlock•5/7/25, 12:57 AM

(Sounds like a lot more than 450 per 10s)

A

Austin BOP•5/7/25, 2:46 AM

@Matt Silverlock I'm not sure which specific limit you're referring to, but as far as the workflow instance creation itself, I think I'm in the clear. I'm using

createBatch

createBatch

to do 100 at a time within each of the initial 31. The total amount gets queued up very quickly. The graph above is the steps processing inside those from my understanding.

I don't see anything about 450 per 10s in the link you mentioned. I've looked through the docs for Workers limits and Workflows limits, and nothing is standing out to me as obvious that my setup would be hitting. Admittedly, I'm not sure my understanding of subrequests and how steps within workflows translate to the workers limits is totally correct.

I was hoping to see some evidence within CloudFlare's dashboard itself as to what is being limited. But I can't find any errors or metrics indicating what limit might be getting hit.

AAustin B @Matt Silverlock I'm not sure which specific limit you're referring to, but as f...

M

Matt Silverlock•5/7/25, 2:50 AM

Sorry, I mean 100 / sec. 31 x 100 per request within a few seconds exceeds that? If not, can you expand on your creation rate? 31 requests over what period? Why 31?

A

Austin BOP•5/7/25, 2:54 AM

@Matt Silverlock 31 is arbitrary and data driven. Iterating over some entities and firing off workflows for related child entities. My understanding is that a single createBatch only counts as 1 request towards the rate limit. I'm getting this from these docs: https://developers.cloudflare.com/workflows/build/workers-api/#createbatch

So each of those 31 is not doing more than 7 createBatch calls (each batch with up to 100). Unless I'm misinterpreting the docs, this seems well within that limit. Also to be clear I'm not seeing issues with the workflow instances being created slowly. It is once they are actually queued, how quickly they get processed. The chart I showed above is not workflow instance creation

AAustin B @Matt Silverlock 31 is arbitrary and data driven. Iterating over some entities a...

M

Matt Silverlock•5/7/25, 3:02 AM

createBatch would limit that. Are you hitting all of this from a single IP? I’d open a Support ticket as hard to dive into over Discord.

I

Isak•5/7/25, 9:58 PM

Do workflows respect smart placement?

MMatt Silverlock createBatch would limit that. Are you hitting all of this from a single IP? I’d ...

O

Ollie•5/8/25, 2:42 PM

@Matt Silverlock how long does it usually take for support tickets to be seen, my P0 one has been opened for 7 days with no response or anything

I

Ibby | Fk Israel•5/8/25, 3:36 PM

Yeah Matt. Is cloudflare looking to hire folks who will help out devs onboard or just continue loosing business ?

IIbby | Fk Israel Yeah Matt. Is cloudflare looking to hire folks who will help out devs onboard or...

M

Matt Silverlock•5/8/25, 3:46 PM

We have a thread open with Ollie here with the Eng lead for the product, so...

IIbby | Fk Israel Yeah Matt. Is cloudflare looking to hire folks who will help out devs onboard or...

O

Ollie•5/8/25, 3:48 PM

Yeah to give them credit, CF have been extremely helpful, and we are greateful that we can talk to eng leads directly

I

Ibby | Fk Israel•5/8/25, 3:48 PM

Sorry i'm just busting balls

M

Matt Silverlock•5/8/25, 3:48 PM

(To be clear, we want this fixed + faster, but hitting some rough edges that we hadn't seen from other teams/customers)

I

Ibby | Fk Israel•5/8/25, 3:50 PM

Tbh; I'm trynna deploy my Nextjs app on cloudflare workers, migrating from vercel.
Tried using the opennextjs-cloudflare but running into build errors, not sure if you guys can help out.

I

Ibby | Fk Israel•5/8/25, 3:51 PM

general-helpMigrate From Vercel - NextJS App Router + API's

J

juanferreras•5/8/25, 11:27 PM

any inherit incompatibility between Workers Rate Limiting e.g.

ratelimit

ratelimit

and a workflow worker?

[[unsafe.bindings]]
name = "RATE_LIMIT_MEASUREMENT_GUEST"
type = "ratelimit"
namespace_id = "1000"
simple = { limit = 3, period = 60 }

[[unsafe.bindings]]
name = "RATE_LIMIT_MEASUREMENT_GUEST"
type = "ratelimit"
namespace_id = "1000"
simple = { limit = 3, period = 60 }

[[unsafe.bindings]]
name = "RATE_LIMIT_MEASUREMENT_GUEST"
type = "ratelimit"
namespace_id = "1000"
simple = { limit = 3, period = 60 }

[[unsafe.bindings]]
name = "RATE_LIMIT_MEASUREMENT_GUEST"
type = "ratelimit"
namespace_id = "1000"
simple = { limit = 3, period = 60 }

export default class WorkflowsService extends WorkerEntrypoint<CloudflareEnv> {
  // ...
  async createWorkflow(payload: Params) {
    // check if the user has exceeded the limit
    const RATE_LIMITER = this.env.RATE_LIMIT_MEASUREMENT_GUEST;
    const { success } = await RATE_LIMITER.limit({
      key: payload.userIdOrIpAddress,
    });
    // success locally works as expected but on production I can x3/x4 the limit and still success=true (all sending the exact same

export default class WorkflowsService extends WorkerEntrypoint<CloudflareEnv> {
  // ...
  async createWorkflow(payload: Params) {
    // check if the user has exceeded the limit
    const RATE_LIMITER = this.env.RATE_LIMIT_MEASUREMENT_GUEST;
    const { success } = await RATE_LIMITER.limit({
      key: payload.userIdOrIpAddress,
    });
    // success locally works as expected but on production I can x3/x4 the limit and still success=true (all sending the exact same

export default class WorkflowsService extends WorkerEntrypoint<CloudflareEnv> {
  // ...
  async createWorkflow(payload: Params) {
    // check if the user has exceeded the limit
    const RATE_LIMITER = this.env.RATE_LIMIT_MEASUREMENT_GUEST;
    const { success } = await RATE_LIMITER.limit({
      key: payload.userIdOrIpAddress,
    });
    // success locally works as expected but on production I can x3/x4 the limit and still success=true (all sending the exact same

export default class WorkflowsService extends WorkerEntrypoint<CloudflareEnv> {
  // ...
  async createWorkflow(payload: Params) {
    // check if the user has exceeded the limit
    const RATE_LIMITER = this.env.RATE_LIMIT_MEASUREMENT_GUEST;
    const { success } = await RATE_LIMITER.limit({
      key: payload.userIdOrIpAddress,
    });
    // success locally works as expected but on production I can x3/x4 the limit and still success=true (all sending the exact same

J

juanferreras•5/8/25, 11:31 PM

I guess if not it could be related to the workflow-consumers (two different apps, but both call it using RPC/service bindings, and they are using

smart

smart

placement)

M

Magic Greg•5/9/25, 4:38 PM

hello - is there a way to terminate or restart a workflow right now?

MMagic Greg hello - is there a way to terminate or restart a workflow right now?

M

Matt Silverlock•5/9/25, 6:23 PM

https://developers.cloudflare.com/workflows/build/workers-api/#terminate

Cloudflare Docs

Workers API

This guide details the Workflows API within Cloudflare Workers, including methods, types, and usage examples.

MMatt Silverlock https://developers.cloudflare.com/workflows/build/workers-api/#terminate

M

Magic Greg•5/9/25, 6:54 PM

Have you tried it? It says it’s not yet implemented when I try it.

森

森@VerseLink (RW)•5/9/25, 8:17 PM

does the waitForEvent count towards the workflow concurrent limit?

J

jason•5/10/25, 7:44 PM

Request for better error messages on workflow runs.

I see constant error messages (red status) on my workflow runs, even when all steps have worked successfully as I expected of them, but the status message doesn't help me know why.

In an effort to get rid of all these red status, I've stripped my workflow down to the bare minimum, and tried over and over...but still get the error in 95% of cases. I'm starting to conclude it's just Workflows being flaky for the end result status and not surfacing a useful error message:

    // Something in one of these two steps causes the workflow to show outcome of "exception" and
    // level of "error". What?
    const { foo, bar } = await step.do("test", async () => {
      const foo = "hello";
      const bar = "world";
      return {
        foo,
        bar,
      };
    });
    await step.do("test", async () => {
      return foo + bar;
    });

    // This will not cause an error...most the time...when it's the *only* step in the workflow.
    await step.do("test", async () => {
      return "foooo";
    });

    // Something in one of these two steps causes the workflow to show outcome of "exception" and
    // level of "error". What?
    const { foo, bar } = await step.do("test", async () => {
      const foo = "hello";
      const bar = "world";
      return {
        foo,
        bar,
      };
    });
    await step.do("test", async () => {
      return foo + bar;
    });

    // This will not cause an error...most the time...when it's the *only* step in the workflow.
    await step.do("test", async () => {
      return "foooo";
    });

    // Something in one of these two steps causes the workflow to show outcome of "exception" and
    // level of "error". What?
    const { foo, bar } = await step.do("test", async () => {
      const foo = "hello";
      const bar = "world";
      return {
        foo,
        bar,
      };
    });
    await step.do("test", async () => {
      return foo + bar;
    });

    // This will not cause an error...most the time...when it's the *only* step in the workflow.
    await step.do("test", async () => {
      return "foooo";
    });

    // Something in one of these two steps causes the workflow to show outcome of "exception" and
    // level of "error". What?
    const { foo, bar } = await step.do("test", async () => {
      const foo = "hello";
      const bar = "world";
      return {
        foo,
        bar,
      };
    });
    await step.do("test", async () => {
      return foo + bar;
    });

    // This will not cause an error...most the time...when it's the *only* step in the workflow.
    await step.do("test", async () => {
      return "foooo";
    });

Here is an example request ID that gives an error status for the above steps inside a workflow: IQI88BKIF6L9BMLR

Anybody else seeing this or know what's going on?

森

森@VerseLink (RW)•5/10/25, 10:26 PM

when i run workflow locally, occasionally i get this error

workerd/jsg/setup.c++:39: fatal: V8 fatal error; location = :0; message = Missing deoptimization information for OptimizedJSFrame::Summarize.

workerd/jsg/setup.c++:39: fatal: V8 fatal error; location = :0; message = Missing deoptimization information for OptimizedJSFrame::Summarize.

and it just crashes the runtime outright (the command shell would be working), but no http requests can be proccessed

森

森@VerseLink (RW)•5/10/25, 10:26 PM

is there a reason why?

T

tyx•5/11/25, 4:23 PM

anyone else getting auth issues when trying to publish worker with workflows?

Getting User settings...
👋 You are logged in with an OAuth Token, associated with the email email@gmail.com.
┌──────────────────────────────────┬──────────────────────────────────┐
│ Account Name                     │ Account ID                       │
├──────────────────────────────────┼──────────────────────────────────┤
│ myemail@gmail.com's Account      │ xbc                              │
└──────────────────────────────────┴──────────────────────────────────┘
🔓 Token Permissions: If scopes are missing, you may need to logout and re-login.
Scope (Access)
- account (read)
- user (read)
- workers (write)
- workers_kv (write)
- workers_routes (write)
- workers_scripts (write)
- workers_tail (read)
- d1 (write)
- pages (write)
- zone (read)
- ssl_certs (write)
- ai (write)
- queues (write)
- pipelines (write)
- secrets_store (write)
- offline_access
🎢 Membership roles in "email@gmail.com's Account": Contact account super admin to change your permissions.
- Super Administrator - All Privileges
error: Recipe `deploy` failed on line 10 with exit code 1
`

Getting User settings...
👋 You are logged in with an OAuth Token, associated with the email email@gmail.com.
┌──────────────────────────────────┬──────────────────────────────────┐
│ Account Name                     │ Account ID                       │
├──────────────────────────────────┼──────────────────────────────────┤
│ myemail@gmail.com's Account      │ xbc                              │
└──────────────────────────────────┴──────────────────────────────────┘
🔓 Token Permissions: If scopes are missing, you may need to logout and re-login.
Scope (Access)
- account (read)
- user (read)
- workers (write)
- workers_kv (write)
- workers_routes (write)
- workers_scripts (write)
- workers_tail (read)
- d1 (write)
- pages (write)
- zone (read)
- ssl_certs (write)
- ai (write)
- queues (write)
- pipelines (write)
- secrets_store (write)
- offline_access
🎢 Membership roles in "email@gmail.com's Account": Contact account super admin to change your permissions.
- Super Administrator - All Privileges
error: Recipe `deploy` failed on line 10 with exit code 1
`

Getting User settings...
👋 You are logged in with an OAuth Token, associated with the email email@gmail.com.
┌──────────────────────────────────┬──────────────────────────────────┐
│ Account Name                     │ Account ID                       │
├──────────────────────────────────┼──────────────────────────────────┤
│ myemail@gmail.com's Account      │ xbc                              │
└──────────────────────────────────┴──────────────────────────────────┘
🔓 Token Permissions: If scopes are missing, you may need to logout and re-login.
Scope (Access)
- account (read)
- user (read)
- workers (write)
- workers_kv (write)
- workers_routes (write)
- workers_scripts (write)
- workers_tail (read)
- d1 (write)
- pages (write)
- zone (read)
- ssl_certs (write)
- ai (write)
- queues (write)
- pipelines (write)
- secrets_store (write)
- offline_access
🎢 Membership roles in "email@gmail.com's Account": Contact account super admin to change your permissions.
- Super Administrator - All Privileges
error: Recipe `deploy` failed on line 10 with exit code 1
`

Getting User settings...
👋 You are logged in with an OAuth Token, associated with the email email@gmail.com.
┌──────────────────────────────────┬──────────────────────────────────┐
│ Account Name                     │ Account ID                       │
├──────────────────────────────────┼──────────────────────────────────┤
│ myemail@gmail.com's Account      │ xbc                              │
└──────────────────────────────────┴──────────────────────────────────┘
🔓 Token Permissions: If scopes are missing, you may need to logout and re-login.
Scope (Access)
- account (read)
- user (read)
- workers (write)
- workers_kv (write)
- workers_routes (write)
- workers_scripts (write)
- workers_tail (read)
- d1 (write)
- pages (write)
- zone (read)
- ssl_certs (write)
- ai (write)
- queues (write)
- pipelines (write)
- secrets_store (write)
- offline_access
🎢 Membership roles in "email@gmail.com's Account": Contact account super admin to change your permissions.
- Super Administrator - All Privileges
error: Recipe `deploy` failed on line 10 with exit code 1
`

I am just using basic wrangler login and this was working till yesterday?

T

tyx•5/11/25, 4:30 PM

funny, the same command re-tried just now works. some glitch

Ttyx funny, the same command re-tried just now works. some glitch

T

tyx•5/11/25, 7:36 PM

this issue is recurring.
fix: just keep trying the same command again and again till it succeeds.

not ideal fix ofc

Ttyx this issue is recurring. fix: just keep trying the same command again and again ...

B

boop•5/11/25, 7:46 PM

Have you tried reauthing?

T

Tobias•5/12/25, 6:20 AM

Is it possible to set concurrency limits on a workflow?
Kind of like how Trigger.dev does it (https://trigger.dev/docs/queue-concurrency).

My use case is to use browser rendering in a workflow and not allow more than 1 run at a time for this workflow in order to not go above the limitations of the browser rendering product.

Trigger.dev

Welcome to the Trigger.dev docs - Trigger.dev

Find all the resources and guides you need to get started

Original message was deleted

T

Tobias•5/12/25, 12:44 PM

How soon is :soontm:? Weeks, months, years?

Ttyx I am getting a very confusing bug. I am getting. `Error: (workflow.not_found) Pr...

Z

zhawtof•5/12/25, 3:17 PM

Now I'm running into this as well

(workflow.not_found) Provided Workflow does not exist

(workflow.not_found) Provided Workflow does not exist

in production.

Anyone on the Cloudflare team know what could be happening? Seems a few other people have run into this.

D

ddsgadget•5/12/25, 11:22 PM

I really like workflows, but just want to say that they are extremely difficult to debug, because the console.logs are not available anywhere. So all you see is the result of a step and have no insight into potential issues when something in the step goes wrong.

Jjason Request for better error messages on workflow runs. I see constant error messag...

D

ddsgadget•5/12/25, 11:27 PM

You are not alone. I've posted this issue on many ocassions. Workflow can run fine, but will show errors in the runs in the dashboard. Someone the other day commented that you must JSON stringify the response from each step and then JSON Parse it back in next step, if you want to get rid of these errors. I haven't tried it yet, and none of Cloudflare examples do this, but apparently that solves these errors.

Dddsgadget I really like workflows, but just want to say that they are extremely difficult ...

S

SakuraFC•5/13/25, 2:54 AM

Aren’t the logs available under the worker that instantiated the workflow? I think I was able to see them (scratches head and trying to remember)

Ttyx anyone else getting auth issues when trying to publish worker with workflows? `...

O

Ollie•5/13/25, 11:19 AM

You might be suffering similar to what we have in this ongoing thread: workflowsI am not sure what has happened with

S

scook•5/14/25, 12:01 AM

Are there any limitations for local development on workflows that might affect the accuracy of

env.WORKFLOW.get(id)

env.WORKFLOW.get(id)

?

After I invoke a long-running workflow in one codepath, a different codepath it sometimes cannot find that instance (i.e. I get

Error: instance.not_found

Error: instance.not_found

). I've been unable to find a pattern for repro, except that it seems the codepath that invokes the workflow can successfully call

get(id)

get(id)

later, even in a different instance of the worker; whereas some other codepath will always get

instance.not_found

instance.not_found

. I'm using the same binding for the same worker with the same workflow id.

J

johtso•5/14/25, 7:34 AM

I see one of the workflows examples shows waiting for a web hook event. How would you deal with the scenario of a web hook being received before the workflow begins waiting for it? Are events queued up in any way?

J

Jonathan•5/14/25, 1:15 PM

Are there any best practices for implementing a workflow that is used for starting and stopping third party services?

I was considering storing a lock in a durable object that states if the start workflow has already been executed you can only run the stop workflow but you still need to track the workflow instance id.

Aalgads I am running into an error for a step: Storage failure for e8202eb85a3fa02c5c2d9...

O

Ollie•5/14/25, 5:24 PM

Limits can be seen here: https://developers.cloudflare.com/workflows/reference/limits/, so maybe something like limit on persisted output from each step

J

johtso•5/14/25, 6:26 PM

Why do the docs say there is a 30 Retention limit for completed Workflow state.. but then also say it's stored for 7 days?

J

johtso•5/14/25, 6:38 PM

Is it not possible to get workflow logs or even see what step a workflow is currently on through the workers API, only the http API?

A

Avi•5/14/25, 7:05 PM

is

wrangler types

wrangler types

wrangler types

wrangler types expected to generate types for workflows defined in

wrangler.toml

wrangler.toml

?

A

Avi•5/14/25, 7:47 PM

ah i see, it doesn't support per-env, but does support toplevel definition

A

Avi•5/14/25, 7:48 PM

also, what happens if i reuse a workflow ID? i know i'm not supposed to but...

AAvi also, what happens if i reuse a workflow ID? i know i'm not supposed to but...

O

Ollie•5/15/25, 7:46 AM

IDs must be unique for a workflow instance, so it will throw an error

Jjohtso Why do the docs say there is a 30 Retention limit for completed Workflow state.....

J

johtso•5/15/25, 2:11 PM

hmm, is this because being able to adjust retention for a workflow is "coming soon" so it's currently fixed at 7 days?

C

cg•5/15/25, 4:42 PM

Hi all!
I'm intersted in setting something up with Workflows, but I noticed on this page:
https://developers.cloudflare.com/workflows/build/trigger-workflows/

That:

Known issue: Restarting a Workflow via the restart() method is not currently supported and will throw an exception (error).

Known issue: Restarting a Workflow via the restart() method is not currently supported and will throw an exception (error).

Known issue: Restarting a Workflow via the restart() method is not currently supported and will throw an exception (error).

Known issue: Restarting a Workflow via the restart() method is not currently supported and will throw an exception (error).

Is this still true? It seems the page was last updated Feb 18 2025, but Workflows has been GA for a month or so now. Can anyone confirm if restart() still throws errors?

C

cg•5/15/25, 4:46 PM

Additionally, does

restart()

restart()

restart()

restart()'ing an instance count to the 100 per 10 second liimit of the Maximum Workflow instance creation rate?

I've recently built a solution on Workflows, and I'm trying to identity if I'm hitting some sort of

Similar Threads