Every time we end a workflow "early", e.g. just return a value before all the steps are done, this i

Every time we end a workflow "early", e.g. just return a value before all the steps are done, this is logged as an Error in CF observability/logs section (even though it is definitely not an error). Is this a known issue? Will it be addressed?

Wilco•6/30/25, 7:40 AM

Hi all, I'm trying to catch an error from a step (when the retry limit is reached) and it seems the causecause of the error is not set. Is that expected? Makes it hard to communicate data from the failed step to the recovery flow

Olriko•6/30/25, 9:50 AM

Hello guys

I'm facing a problem where my workflow instances get queued but never starts. After a mistake that end up calling instances recursively.

OOlriko Hello guys I'm facing a problem where my workflow instances get queued but neve...

avenceslau•6/30/25, 9:51 AM

Please link your account so that we can have look

Aavenceslau Please link your account so that we can have look

Olriko•6/30/25, 9:54 AM

https://dash.cloudflare.com/e97e3fed722428f4d726062eae9aa766/workers/workflows

SODaLC•6/30/25, 12:36 PM

Is there a way to make Workflows process one at a time, instead of multiple running at once?

MithushanJ•6/30/25, 2:10 PM

What is the time limit for a step to finish in a workflow that has multiple steps

alex (he/him)•6/30/25, 6:07 PM

Any way we can get step.waitForEventstep.waitForEvent not count towards the total step count per workflow, similar to step.sleepstep.sleep

alex (he/him)•6/30/25, 6:08 PM

We have a hard time to strucutre our workflows to stay under the 1024 step limit. We either hit the cpu time limit of 5 minutes if we do multiple tasks in one step, or we hit the step limit if we do each task in a separate nested workflow

Gabriel Francisco•6/30/25, 6:52 PM

how can I see the logs of a specific instance? I can't find it on observability

alex (he/him)•6/30/25, 9:33 PM

Are memory limit of 128MB shared between multiple workflow instances?

alex (he/him)•7/1/25, 6:24 AM

What is the backoff rate from exponentialexponential is it 2?

Aac Every time we end a workflow "early", e.g. just return a value before all the st...

ddsgadget•7/1/25, 11:12 PM

This has been an issue forever. I have posted about it many times and others have experienced the same thing. I have never seen Cloudflare acknowledge it, though. Supposedly, if you JSON stringify the response, you won't see an error, but even Cloudflare doesn't do that in any of the examples. Bottom line, this is a bug in workflow, but it doesnb't seem like they will ever fix it.

Dddsgadget This has been an issue forever. I have posted about it many times and others hav...

Seekerdasbatatas•7/2/25, 9:01 AM

While the invocation might show the invocation as errored/canceled, it's not directly related if your workflow instance threw or not, it's a side-effect of how we guarantee durable execution internally. You should refer to the instance logs (in the Workflows tab, in dash),to see if an instance failed or not.

But this is something that we should definitely fix in our logging pipeline, so sorry about that :MeowHeartCloudflare: , we'll try to get it fixed :soontm:

SSeekerdasbatatas While the invocation might show the invocation as errored/canceled, it's not dir...

acOP•7/2/25, 12:18 PM

Thank you! That'd be hugely helpful for us, especially since it makes it show on our dashboard that we have hundreds of thousands of "Errors" (most of which are actually just workflows). Makes it very hard to understand our system health.

Olriko•7/3/25, 11:36 AM

Hello, I have a 429 error when I want to trigger a workflow from the ui (and I don't spam your api)

POST https://dash.cloudflare.com/api/v4/accounts/e97e3fed722428f4d726062eae9aa766/workflows/workflow-orchestrator-staging/instances

POST https://dash.cloudflare.com/api/v4/accounts/e97e3fed722428f4d726062eae9aa766/workflows/workflow-orchestrator-staging/instances

Mm0c What confused me, I think they only start showing up on observability when they ...

unsync•7/3/25, 4:19 PM

We started on workflows and this puzzles me, if I have a long running workflow, I get close to zero observability of what happen inside my steps until the whole execution is done ?

martinhoest•7/3/25, 7:43 PM

I must be fooling about with Workflows and Observability - or is it normal that a failing step of a Workflow will not add meaningful logs to Observability?

The error here is thrown within step.dostep.do but it is nowhere to be found from the logs tab of the worker. I also cannot find anything on the Instance ID of the workflow invocation.

SSeekerdasbatatas While the invocation might show the invocation as errored/canceled, it's not dir...

ddsgadget•7/4/25, 2:17 AM

Thank you! yeah, I understand the error has no impact on the actual workflow. But, it is really annoying to see the errors in the dashboard when there is no real workflow error.

Gabriel Francisco•7/4/25, 3:46 PM

is it possible to see the logs of a particular step? this seems basic but I can't find a specific filter for that

Gabriel Francisco•7/4/25, 4:37 PM

my workflows are going to some weird "Waiting" state, even without sleeps, I would like to see what is happening but can't find any logs

martinhoest•7/5/25, 6:45 AM

What's the expected output in logs if one Workflow triggers another Workflow, but the instance ID of the latter already exists?

Will I need to loop through all created instances from await this.env.WF_FILING_INGEST.createBatch(batchParams);await this.env.WF_FILING_INGEST.createBatch(batchParams); and call their status()status() function?

Or it might be better to check for existing instance before creating new?

Bryan Chen•7/7/25, 9:52 PM

what exactly does this mean? the workflow status is completed, but the step is running and have no output?

BBryan Chen what exactly does this mean? the workflow status is completed, but the step is r...

johtso•7/9/25, 10:03 AM

Did you figure out what was going on?

Bryan Chen•7/9/25, 8:50 PM

no. I suspect the network request got hang but wasn't able to confirm it. and that still indicate some bug in the workflow

Advany•7/9/25, 9:51 PM

Is there already a way to use workflows using vitest?

https://github.com/cloudflare/workers-sdk/issues/7414

GitHub

🐛 BUG: Vitest does not work with workflows · Issue #7414 · clo...

Which Cloudflare product(s) does this pertain to? Miniflare, Other What version(s) of the tool(s) are you using? 0.5.32 [@cloudflare/vitest-pool-workers], 3.91.0 [wrangler] What version of Node are...

BBryan Chen what exactly does this mean? the workflow status is completed, but the step is r...

Jadugar_Jaggu•7/10/25, 3:58 AM

Maybe you haven't used await in your step. it's waiting for pending promise.

try awaiting every step "await step.do()"

JJadugar_Jaggu Maybe you haven't used await in your step. it's waiting for pending promise. tr...

Bryan Chen•7/10/25, 4:09 AM

I global searched it, every step are awaited. besides, if missing await can trigger this, it is a workflow bug

Bryan Chen•7/10/25, 4:13 AM

related question, the step timeout is 10mins, is it wall time or compute time?

AshuRajput02•7/10/25, 9:22 AM

hey everyone
so there is a site of mine say abc.com now i m thinking to apply zero trust policies in it . so basically i want ki user should be able to access the auth page only via warp client and not by directly hitting the url on website.
how to achieve this ?/

Jeffrey Konowitch•7/10/25, 11:22 PM

Clarifying Maximum Workflow instance creation rateMaximum Workflow instance creation rate Limit

The createBatch documentation says

call to createBatch is treated the same as a call to create (for a single instance) and allows you to work within the instance creation limit.

Does this imply that when using createBatchcreateBatch , if you were to create 100 instances per batch (the maximum), that you could create 10,000 instances (100 per batch * 100 calls) per 10 seconds?

I understand there are other limits that might kick in (e.g. Maximum number of queued instances), but this example is meant to clarify the creation rate limit.

Worklows Limits Reference

Frank•7/11/25, 4:03 PM

trying to move one of our cache gen into workflow

list active users
fetch data
store into r2

my initial thought is having a workflow just to fetch users, then it kicks off another workflow to fetch and cache data based on user id
or better to just have a single workflow?

flof.fly•7/12/25, 9:54 AM

Can I invoke a Container from a Workflow?

flof.fly•7/12/25, 11:49 AM

Yes I can

Joentze•7/19/25, 3:54 AM

Seeking advice here, i want to create scheduled workflows. I want to create periodic tasks (i.e. daily at 9am, 1st day monthly), and one-time tasks (date xx at yy time). These workflows are dynamically added by the user via an api request to the worker. I’m using Hono to handle routing.

What should i use to achieve this?

JJoentze Seeking advice here, i want to create scheduled workflows. I want to create peri...

boop•7/19/25, 11:13 AM

Durable Object alarms are better suited for this

Yash Gupta•7/21/25, 9:29 AM

Has anyone tried to access steps/execution history for workflows from within their application?

It seems that the workers binding for workflows only gives present status and output, and while the API itself has steps history, it doesn't work for local development, making it hard to incorporate that as part of the solution.

Another option I've explored is manually intrumenting the step, waitForEvent and sleep methods, but it seems like redundant work (and possibly flaky)

alex (he/him)•7/22/25, 11:44 AM

I'm really having some issues with workflow creations

alex (he/him)•7/22/25, 11:44 AM

Already using createBatchcreateBatch and try to create ~650 instances all at the same time

alex (he/him)•7/22/25, 11:45 AM

That would be 7 calls with createBatchcreateBatch.. however I'm still seeing super long time until createBatchcreateBatch resolves as well as a lot of too many requeststoo many requests errors.

alex (he/him)•7/22/25, 11:45 AM

This is super hard to debug and annoying too.. why does even createBatchcreateBatch limits soo hard?

Bboop Durable Object alarms are better suited for this

Rahul gangotri•7/22/25, 1:09 PM

alarms will be expensive as it also charges on wall time, so if a task is taking longer to resolve or making call to external api is in place, then its going to cost for all that wait time.

acOP•7/22/25, 2:08 PM

How likely is it that the "Maximum Workflow instance creation rate (100 per 10 seconds)" limit will be increased? This limit is pretty low for us

RRahul gangotri alarms will be expensive as it also charges on wall time, so if a task is taking...

boop•7/22/25, 2:20 PM

Make the DO call a workflow, just use the DO alarms as a scheduler

Aac How likely is it that the "Maximum Workflow instance creation rate (100 per 10 s...

alex (he/him)•7/22/25, 3:00 PM

This limit is pretty low for us

It's the same for us. We are already looking at alternatives to Cloudflare Workflows, since it doesn't work for high load use cases.

Aalex (he/him)> This limit is pretty low for us It's the same for us. We are already looking ...

Jeffrey Konowitch•7/22/25, 3:46 PM

I've asked about this three times and no response from Cloudflare. A bit concerning for anyone looking to build apps with even moderate throughput requirements.

workflows
Clarifying `Maximum Workflow instance creation rate` Limit
Is the workflow creation rate limit 100 per 10 seconds for createBatch?

hype•7/22/25, 4:07 PM

Same. Just running into this when trying to migrate

hype•7/22/25, 4:07 PM

and now looking again into alternatives.

Hhype and now looking again into alternatives.

alex (he/him)•7/22/25, 4:32 PM

There are not too many for our usage pattern. Also noone can really beat Cloudflare's pricing.

Every time we end a workflow "early", e.g. just return a value before all the steps are done, this i

Similar Threads

Similar Threads

Similar Threads