I’m running into something strange and wanted to sanity-check with the community.
I’ve been using Bubble’s API Connector to send POST requests to Stripe (e.g., creating customers). My setup was working perfectly before using:
Data type: JSON
Body type: Raw
Body:
email=_*_email_*_
Now, suddenly — all my Stripe calls have stopped working, including previously working ones. According to Stripe logs, Bubble is now sending the literal dynamic placeholder string, like _*_email_*_, instead of replacing it with the actual value. This causes formatting errors when Stripe receives the request.
It seems like Bubble is no longer parsing dynamic parameters properly in Raw body mode — and it behaves the same with any param wrapped in _*_param_*_.
I’ve attached a screenshot showing a simple Bubble API call setup, as well as a screenshot of Stripe’s response in the logs.
A few quick questions:
Has Bubble recently changed how dynamic parameters are parsed in Raw or Form-data body types?
Is there now a safer or more reliable way to pass dynamic values to Stripe via the API Connector?
Has anyone else seen this behavior start recently?
It seems to be a problem with the bubble, I’m having the same problem here, it seems that the dynamic values in the raw type in the api connector aren’t working.
I first noticed the issue around 2:50 PM EST, and as of 6:50 PM EST, everything is suddenly working again. Bubble is now correctly replacing dynamic parameters like _*_email_*_ in Raw body requests as expected. All my Stripe API calls are back to normal and behaving just like they were before.
It seems like this may have been a temporary glitch or something quietly patched on Bubble’s end.
Would love to know if anyone else experienced the same issue — and if someone from the Bubble team could shed some light on what might’ve happened, that’d be helpful!
Also experiencing this issue. I tried the workaround above, but it didn’t solve the problem. Weirdly, the issue means it only fills in the first parameter and not any others.
Reported it to support and they’ve said they’re looking into it. Another workaround I tried that has got it back up and running is creating a duplicate API call and just having a single _*_body_*_ parameter and formatting the JSON body with the additional parameters myself in an arbitrary text, ensuring that strings are then formatted properly using formatted as JSON-safe. Not ideal, but got things working again.
Confirming same issue as discovered by @gavin3. In my case we are using Microsoft Oauth/Token endpoint and raw format is the only accepted body. Hope this will be resolved soon.
Im having same issue. our entire marketplace is broken today as a result. have emailed bubble and hoping for an immediate fix on this before we lose all of our subscribers. Connector is passing the labels instead of the values as detailed in above user examples. This MUST be fixed today or we are completely screwed. heres our call body:
Stripe is just getting passed the dynamic value indicator/label and not the dynamic values so of course all 700 of our subscribers websites are now failing to load the checkout page when taking an esim order.
I cant even fathom why this would happen and not be announced by bubble or immediately repaired. Literally entire platform is down now
Note, the brtowser seems to be ignoring my _ symbols in this post also but it should show underscore*underscore
I honestly cant believe we are in this situation today. Like we are going to get sued by our entire user base over this if its not resolved asap and i cant even call bubble to ask whether they have picked up our support ticket on the matter. This is a huge technical failure on bubbles part!
This is crazy, how does something like this even happen? What are they doing to the API connector that this could possibly happen?
Edit: I looked through my calls and everything seems to be working fine, could be because all of my dynamic parameters have zero underscores or spaces?
Hello,
I’m Zoe, an engineering manager at Bubble. Thank you for flagging this. We have deployed a fix for this issue. Please let us know if you are still experiencing any unexpected behavior. We apologize for the inconvenience this has caused.
Yikes - bugs like this are scary for production apps where we need reliability and full visibility into our apps. Especially when there is no easy way to find if this caused issues in our workflows and data (by the sounds of what others have said about this bug this can vary depending on the API call)
So then one day in a few weeks a user of our app might complain about dirty/disconnected data caused from this, that then we need to try and debug we can’t find the issue because the logs are too old to query or don’t show errors. Days of wasted debugging work and lost trust with users just to try and find this anomaly. Or worse we never know an issue happened but our users see the app’s data is incorrect and lose trust or churn without ever reporting an issue.
I would take a Bubble wide outage over bugs like these any day
If the endpoint we are calling wasn’t our own and I would not have access to the logs of it, this would have been almost impossible to debug. The bubble logs for alle API calls looked fine in the meantime…
@zoe1 can we please get some clarity from the bubble team regarding:
The exact duration that this issue occurred
The specific set up of calls that may have failed
As others have pointed out, this would be an incredibly difficult problem to debug. We need to know which calls may have been affected and for how long so we can assess any potential damage. “Some Api Connector calls” isn’t specific enough.
