Yes also seeing this. Both in Test and in Production @aj11 & @josh .
It’s killed the performance of our platform. Any updates to data take 3 seconds+ to register - this was instantaneous previously. My assumption this is the same issue.
Is this affecting apps Bubble-wide or is it localised?
The issue is so severe that I would have expected more people with production apps to be on this thread by now…
these update bugs are making our app useless…
@josh If you look at the Gmail app, the list of messages does usually auto-update, but they also have “pull-down-to-refresh” to force an update from the database if needed. This is pretty standard operating procedure for modern apps because you’re never going to be able to auto-update something 100% of the time. There are just too many variables to account for. Even if you get it figured out for now, one change to one browser or caching system could completely break the system again.
I appreciate the work you guys are doing to try and solve the auto-update functionality, but I still strongly believe that a lack of ability to “force refresh” a RG from a workflow is a huge vulnerability for Bubble and anyone that uses it in the long term.
Yes - we have a major pitch tomorrow and this bug has killed performance to the point that I’m not sure that we can even show the platform. That’s a pretty big disaster for us.
On Wednesday everything ran perfectly but by Thursday it had fallen unresponsive. No response from Bubble yet on any channel. This type of episode puts platform stability at the top of our investors concerns which is frustrating as we have really won them over to Bubble.
@aj11 are you still seeing the same behaviour?
I had a chuckle when I stumbled upon this thread [Solved] Stop Repeating Group from Automatically Updating
Looks like Bubble heard him 
This is crippling us. Our theory is that something has changed in the way Bubble works, what previously was cached is not any more. So we are looking at profiling to see what is taking the time.
Like this message if you think adding a “Force Update RG” workflow action would partially improve the situation with this bug by providing a manually triggered workaround.
@aj11 I’ve profiled our application and narrowed our particular problem down to a change in the performance of publishState in plugins - I see what previously took no perceptible time, now takes ~800 milliseconds per publishState 
- so we have 6 publishStates and this takes around 3.2 seconds and that is where out time delay comes from. Probably not what is going in in your case but might help you (or not).
Ouch! Thanks for the heads-up. Might be a good PSA for the plugin forum. (I have another question but don’t want to derail the topic of this thread.) Maybe it even warrants a bug report?
On that daily “refresh the page after every change while demoing new potential users” grind. Giving the old “this is something we’re working has quickly as we can to improve” line.
Any updates on a fix @josh ?
@lindsay_knowcode and I now have a bug report now in progress on this - as in, Bubble have gotten back to us.
We’ll keep this thread updated, or spin it off if the publishState() issue turns out to be something altogether different - though it’s possibly the the same as the impact is most noticeable inside repeating groups where those delays get aggregated for n instances of the element.
Just wanted to highlight this forum post I made after discovering the same issue with publishState for anyone who might be having issues:
Honestly, it looks like two things are going on here:
Echoing @exception-rambler, its unclear if this is the same as OP’s issue, but it seems to have started around the same time (we noticed performance issues starting around the end of February). Just wanted to leave it there for visibility in case it helps anyone.
Can you clear the repeating group and then redisplay it, thereby forcing a refresh?
I’ve tried everything like this and there isn’t currently a way to hack it. Bubble’s call optimization for performance improvements stops all of this from working.
Happened on my app today again. Coincidence or not but it was exactly at the times when the Externally Measured Latency (bubble status) chart data was above normal.
This topic was automatically closed after 14 days. New replies are no longer allowed.
