If you’re having this problem for the first time, you need to file a bug report with the Bubble team so they know you’re experiencing it. The more users they know are affected the faster they will move to fix this.
This is on our radar and we’re looking into it. +1 and thank you to everyone who submitted bug reports, that helps us narrow down the problem.
There’s a system metric we think (but are not sure) is correlated with this problem. Assuming we’re right that this metric is an accurate predictor of the problem, things should have started working around 10:11 am ET (ie, 47 minutes ago). Does that match up with your observations?
Thanks for the update, I started noticing the issue around 8 am ET today.
I was able to reproduce it consistently until approximately 10 am ET so it does match up !
Edit : 12 am ET experiencing the issue again consistently on multiple apps @josh
Edit 2 : 12.19 am ET not able to reproduce the issue
@josh like I mentioned in my previous post and other threads that I have initiated, this is not a recent phenomenon. It happens often in our applications. Our users and I just keep refreshing the pages now whole day. Support team refuses to accept our issues as bugs as usual.
2nd what @mghatiya is saying. This has been an issue for months and slowly getting worse and worse. Been basically gaslighted by support every time I previously reached out since its only reproducible during certain times of day.
This isn’t just an inconvenience, elastic search with “do a search for” is a backbone of Bubble functionality, UIs of complex apps are basically just a static webpages without the ability to auto update lists of data.
This being the case I appreciate your insight into what’s being done, but “looking into it” doesn’t sound like a fix is being prioritized as much as a bug of this caliber warrants.
Note: This behavior is not currently occurring in my app. So thanks for any work you guys did to fix things in the short term, just concerned about this issue returning tomorrow if it is not addressed in a more fundamental way.
We now know the source of the problem; it has to do with one of the systems involved in our real-time update getting backed up for certain kinds of queries. It’s an intermittent problem, and as people have mentioned in this thread, our metrics indicate that it’s been slowly getting worse over the last couple months
We have what we think might be a short-term band-aid fix in place already. We’re not sure it will work; we should know tomorrow morning, which is when based on previous days’ patterns we expect the issue to get worse again
Regardless of whether the short term fix works, we have a long-term fix that we’re pretty confident will completely solve the problem. This may take several days to roll out, because it involves making substantial changes to our systems.
Apologies for the slow response on this; I know intermittent bugs are extremely frustrating to deal with. That said, we now know for sure that there’s a problem here and what the cause is, and will get it fixed as soon as we can.
