Aits a beta... just opened up... lets be kind that they didn't wait till everything was polished 
FOf course, but it's been a few days since it is officially known, they could have added a note that this is currently not supported IMHO, esp. since this is so prominently mentioned in the docs
Ayeah, there are also a couple of other things. I think it will get fixed soon (like the create locally needing an id)
Bis it possible to access env vars from inside workflows? e.g. to access db usually a token/url is needed that is saved in the env
IAccess is the same as a normal worker, but it requires a
this. before accessing it… it’s not mentioned in the docs, they will add itBI see, thank you!
Bin general are there any materials that show cases of proven practices that work well with workflows? i.e. is it better to keep interaction with the db inside the workflow or outside of the workflow in the worker itself. or more generally what actions are better to perform inside workflow vs. inside a regular worker?
AI made a workaround. Add the workflow items to a d1. Then a do sends it every 10 seconds to the workflow. Dripping it into it.
Would be great if we could just add that on step level of workflow level
Would be great if we could just add that on step level of workflow level
AJust though about it now. I could have created a workflow that goes thru the list and sleeps between each entry...
IThe less work each step makes, the better. Each step will be retried, persisting results from other steps, but not variables outside of those (so the variable assigned from a step return is maintained, a random variable is not)
IDo a db step, do a processing step (or even a per item processing, if it’s a loop and it’s not too long to go over the step limit)
IDo consider that you can nest steps, too. If you were to need that.
Bthank you
DIsn’t the step limit 250? So I have a step that return an object of 500 items and need to loop over each item to make another external api call and do further processing, I can’t have a nested step inside the for loop to process each item? Since it sounds like I’d hit the step limit
IExactly, yes. The limit is 255 steps.
IDo remember that you still have the 1k subrequest limit that applies and that retries will be counted against this limit.
DHas multiple workflows with local wrangler been "fixed" yet? I can't see any issues in the
wrangler-sdk repo. Shall I create one if it's not been fixed? Not sure where the state of this bug would be kept
Dstill broken as of last night since i ran into it
DHmm maybe i'll process 10 at a time then
MI haven't seen this mentioned anywhere in the docs - I can't figure out if sleeping executions are counted towards the concurrency limit.
general-helpDo sleeping Workflows count towards the maximum concurrent execution limit?
general-helpDo sleeping Workflows count towards the maximum concurrent execution limit?
SIs there a way to have a monorepo of several workflows in one repository ?
Sis someone succeded with same setup as for workers?
MYes, but right now that is broken in local dev. Search this channel for “multiple Workflows”
SCan I add my workflow definition to my existing hono cloudflare worker that contains all my other api logic, or do I need to set this up as a new project?
MYes, you can add as bindings. Also, you can create an endpoint to manage your workflow by reaching your instance. You can start, pause or resume your workflow via that hono endpoint.
Dhi, trying to add workflows like this in wrangler.toml
i have it under a [[env.staging]]
if i have one i works but with two the worker failes to start
"The Workers runtime failed to start. There is likely additional logging output above."
there is no error in the log. what am i missing?
i have it under a [[env.staging]]
if i have one i works but with two the worker failes to start
"The Workers runtime failed to start. There is likely additional logging output above."
there is no error in the log. what am i missing?
DOk, seems like i posted to fast and there is an answer above. Bug in local dev(?) any timeframe in getting that in dev hands?
FCloudflare does not do ETAs
F:soontm:
Dhaha fair ^^
Sfeel like i must be doing something stupid but trying to run the workflows-starter and getting a 500 when trying to create a workflow:
will thread the rest of my src code. is there a specific url you're supposed to hit or something?
will thread the rest of my src code. is there a specific url you're supposed to hit or something?
SHey guys, within the run function of the workflowentry point, how do you get the current id of the workflow? Is it within this.ctx or something? Thanks.
ZHello, I've redeployed the remix-template with a worker and workflow following the latest remix template available to support Workers deployment instead of Pages. The same error message
Worker "workflows:demg-test-workflow"'s binding "USER_WORKFLOW" refers to service "core:user:" with a named entrypoint "TestWorkflow", but "core:user:" has no such named entrypoint. occurred. I've created a minimal repo to recreate and added the full log message in this issue FYI. https://github.com/micksabox/remix-cloudflare-workflows/issues/1GitHub
wrangler 3.84.0 (update available 3.84.1) ------------------------------------------------------- Running custom build: pnpm build > demg-remix@ build /Users/michaelnolivos/Documents/Projects...
wrangler 3.84.0 (update available 3.84.1) ------------------------------------------------------- Running custom build: pnpm build > demg-remix@ build /Users/michaelnolivos/Documents/Projects...
SAnyone know how to fetch the instance id of the workflow of itself within the run function of the workflow? Thanks.
JBoosting this. I was not able to find a way to programmatically fetch the workflow id either.
JIs there a way to delete a workflow? Used onto to send messages onto a queue and some during testing and want to just clean them up for my OCD 
Djust to clarify, it means that if you have 5k queued (scheduled workflows pass to queued, once they are intended to be run), 100 will be initially instructed to run, and as they complete, more will run?
DThis is not possible for the moment
SAh ok. Well, something you guys can add then.
Ahi! what is the unit of delay field in WorkflowStepConfig if its a number. the documentation at https://developers.cloudflare.com/workflows/build/sleeping-and-retrying/ says default is 10000 - is this ms?
Cloudflare Docs
This guide details how to sleep a Workflow and/or configure retries for a Workflow step.
SHi 
. Does anyone have a demo Remix repo I could learn from? (using remix-run/remix/templates/cloudflare-workers)
I have tried following @Craig's "punderful-workflows" example, and kept it very simple in the wrangler and the "main" JS file (following his example, but in a Remix repo)....
But no matter what I do, I always come up against this error....
If someone has a repo with workflows in a Remix setting, that would be much appreciated
. Does anyone have a demo Remix repo I could learn from? (using remix-run/remix/templates/cloudflare-workers)I have tried following @Craig's "punderful-workflows" example, and kept it very simple in the wrangler and the "main" JS file (following his example, but in a Remix repo)....
But no matter what I do, I always come up against this error....
Worker "workflows:interaction"'s binding "USER_WORKFLOW" refers to service "core:user:" with a named entrypoint "InteractionWorkflow", but "core:user:" has no such named entrypoint.If someone has a repo with workflows in a Remix setting, that would be much appreciated
MYou can’t deploy a Workflow as part of a Pages app (which is likely what the default Remix build is trying to do).
Deploy the Workflow separately (as a Worker) and bind to it from your Pages app / call it over HTTP.
Deploy the Workflow separately (as a Worker) and bind to it from your Pages app / call it over HTTP.
AI've been digging into workflows a lot, but recently I've been getting this error:
The destination execution context for this RPC was canceled while the call was still running.ZHi Matt, the repo that I posted is using the latest Cloudflare Remix template using Workers instead of pages. And the issue was still occuring.
WThe Remix template is currently building as a Pages project then deploying as a Worker which is likely why this is happening. Remix has native support for Assets now so we're moving that over, you can track that here: https://github.com/cloudflare/workers-sdk/pull/7170
SAh thanks for the reply. Can I ask - what determines Pages vs Worker? I was previously looking for "pages_build_output_dir " in .wrangler
DI'm trying to understand how can I get workflow step by workflow id or is not possible?
