Bug with API workflows not properly scheduling other API workflows

[david.j.hansford]

I have already submitted a bug report. Hoping for quicker attention here in the forum.

All this relates (currently) to my development server. I haven’t tried pushing anything live in fear this behaviour will be repeated there too!

Whenever one of my API workflows schedule another API workflow, the second API workflow does not run. Looking at the logs you can see “Schedule [name of API workflow]” with all the variables listed as expected and no mention of condition failure. However, nothing in the second API workflow happens. Every action in the first API workflow occurs as expected other than scheduling other API workflows.

For example:
API workflow: “message_changed”
When I send data to this endpoint from an external source, “message_changed” starts.
Step 1 (schedule another workflow) appears to schedule in the logs (see screenshot attached) but nothing further happens. No conditions are failed.
Step 2 fails its condition and is skipped (as expected).
Step 3 creates a new database entry as expected.

I know there was wider outage this morning but is anyone else experiencing ongoing issues, perhaps similar to this one?

Also if I’m just an idiot and have missed something simple please let me know.

[david.j.hansford]

morning bump. This is a serious issue, really hope someone can help!

This is happening to ALL my workflows across the dev server. Frontend and backend. New and old. Old ones that were working before the bubble crash even.

[david.j.hansford]

Still no response from anyone at bubble. I’ve filed a bug report, requested help from support, posted here and tweeted them.

Update: I’ve tried updating bubble and reverting back to before this started happening. No effect. Now approaching 48 hours of unusable bubble on our paid app with paying customers who rely on what we’ve built for them.

How do I get a response out of bubble? @josh

[mikeloc]

I wouldn’t normally do this, but it does sounds like an extenuating circumstance that you might want to be alerted to, @sam.morgan.

Best of luck getting the matter resolved, @david.j.hansford.

[sam.morgan]

Thanks for the ping, @mikeloc.

@david.j.hansford - I just checked and we have received your bug report - since it was filed at 5:30pm EST yesterday, we’re still getting through some of our overnight backlog (at this time our bug report investigation hours are 9am - 6pm EST), which is why you still haven’t received a response.

It can sometimes take up to 2 hours of active business hours for us to get to a message and respond that we are looking into it. I’ve pinged this to the support rep assigned to your ticket, and we’re investigating now!

Thanks so much for your patience - we’ll follow up via email as soon as we have more information.

[david.j.hansford]

Thanks @mikeloc and @sam.morgan. Really good to get some feedback, hope we can get this one resolved quickly

[david.j.hansford]

Sadly (for me) it turns out that I’m just an idiot.

In case anyone finds this thread because they’ve done the same thing, the solution is simple:
Scheduled workflows were simply paused under Logs > Scheduler > Pause tasks.

I’m going to generously allow myself to believe that this got paused during the bubble outage yesterday but either way I know where to look if something like this happens again. Thanks all!

(Bubble support just got back to me with this solution.)