I’ve been noticing some strange behavior with Bubble’s scheduled API workflows and I’m wondering if anyone else has experienced this.
Sometimes when I schedule an API workflow on a list, the workflow appears as successfully scheduled in the logs, but the actions inside the API workflow never actually run. There’s no obvious error, it just seems to stop after being scheduled.
This has been happening intermittently, especially in cases where I’m using backend workflows to make changes to a large number of users or records. For example:
The workflow shows as scheduled in the server logs
The API workflow itself is triggered
But the actions inside it (such as making changes to users/data) don’t execute consistently
Because of this, some users get updated while others don’t, even though everything looks correctly scheduled on the backend.
I’m trying to figure out:
Has anyone else experienced this issue?
Is this a known limitation or reliability issue with scheduled API workflows on lists?
Are there best practices to make these workflows more reliable when processing large amounts of data?
It is possible for bubble to timeout if too many events are trying to be triggered at the same time. Can you spread out the action by updating the seconds between scheduled on a list of things and see if that helps?
it’s not a lot of things to be honest.
it works for most of the time, I can say 90% or maybe 95% but the remaining 5- 10% creates unnecessary explanation that I have to do for the app owner. and those remaining % matters too.
Yes, I have history logs which lets me know when it runs.
The task end time is not empty, i have confirmed from the database, so i can say to some extent it’s not a problem from my app.
These workflows have quite a lot of actions in them and seemingly no conditionals. Are you certain that later steps are not dependent on previous steps in any way? This action list looks like it could be extremely prone to race conditions or unwanted overwrites (editing List A after List B, etc.).
