I’ve noticed my app’s performance swinging WILDLY today. The site has been live for about 2+ weeks, and the performance has been relatively consistent and snappy.
Today there have been wild swings with no explanation. Pages that typically load in 2 or 3 seconds can randomly take 20 - 30 seconds (essentially “down” as far as users are concerned). Nothing material has changed, and I’ve ruled out the following:
Page Design - no changes have taken place to page design.
Database Calls - all queries are exactly the same.
Usage - light, typically no more than 5 - 10 concurrent at any point, and that’s been relatively consistent, backed up by Google Analytics data.
Browsers / ISPs - when things turn south, it happens across all devices, browsers, and connections I can get my hands on to test.
I don’t have visibility into anything but what I can control, but this appears to be outside of that realm. When things are crawling, it appears to be elements on the page that suffer the most (e.g. images or other items that have “Do Search…” behind them, like pulling up a user’s profile picture to place next to a post).
Finally, my app also has 2 extra CPU units purchased… but with a mere 5 - 10 concurrent users right now as the “norm,” that SHOULD be overkill already.
Bubble - please help! This issue is severe enough that it’s the equivalent of unusable for many users, and I’ve reached out to support already and have gone back and forth a few times explaining this, but I also wanted to see if there was something going on with other Bubble apps out there.
Adding @emmanuel and @josh if there’s any additional light you can shed on this.
I have to say that in the past 2 days I have started to experience similar issues. Mainly with consistency in app reliability. Users started saying that things are not saving for them anymore in some cases which has never happened before no matter how long the load time was. So that is a bit disturbing. I imagine it is about server max capacity and then certain actions timing out but it shouldnt happen in the app’s current state.
About the units, we have even less concurrent users but I just purchased my fourth unit hoping this will smooth out the bumps. We do have a lot of api workflows (on lists) so that takes up quite some capacity.
I guess it will be hard for the bubble team to say anything sensible about this though!
Yeah, I launched on Monday after 9 months of preparation. So far it has been the biggest disaster possible. I have a news release going out tomorrow and have already spent a TON of money on advertising in the past 3 days. Just so people can go to a broken website. I hope bubble is working overnight on this because I am out of business until this is fixed and am looking like a joke to my customers. This is also after purchasing now 2 additional Units while that didn’t look like the issue but was told to buy more from bubble.
Our system metrics today look normal, but I’m double checking now to see if there’s something the metrics are missing. We track things like median page load times across all bubble apps, which generally pick up performance problems that affect most apps, and haven’t gotten any notifications there, but it’s possible that things are breaking in ways that don’t show up in our usual metrics.
We did have a couple outages yesterday (see status.bubble.is for the exact time), but the incidents were pretty brief and didn’t result in long term performance degradation.
Also, keep in mind that if your app’s usage graph is showing any maxed out time at all, that means performance was degraded during that time and more capacity would have helped. If your usage graph does not have any maxed out time, though, more capacity will not help.
I am having issues as well. I am not able to do a simple search and login functionality stopped working for me in prod all of a sudden. The website has become unusable.
