Jici, you’re one of my hopes around here. I’ve been having similar thoughts about the leadership. Ever since they started to push mobile and AI, we have had waaaay more issues than I was used to over the years.
Unfortunately, my decision is made for now and I’m not building more bubble apps. Testing a few solutions already, but still quite sad about it.
Totally agree with you.. but.. I’m more than sure that there is a time window where very few people is online and be able to deploy this breaking changes with minimal impact if something goes wrong. Doing so in peak hours for Europe and the Americas is simply ridiculous.
Dont build here honestly i am putting resources on migrating to more native and reliable technoligies
The problem is the # of request. Even if they have deployed this during the night, the issue would have started to show during peak hours when the # of requests increase. I prefer they did this on a “scheduled maintenance” (but they should have said this was always ongoing…) so we all know this could happen.
bubble is for basic mvp if you ever grow they offer enterprise plans at a price that they will become your partner because they will take all your profits
Its quite rare on dedicated. I remember only one issue related to the images and plugins allocated in one of the shared, but it was never fully down not even once in almost 2 years of dedicated/enterprise.
Main thing was being able to choose when to update to the next Bubble version.
Are regular outages during peak business hours just what we should come to expect with Bubble? These outages are so embarrassing for us as a company. It would be great if Bubble shared our embarrassment and did something to prevent this from being an every 10-14 day issue. Doing this maintenance during prime time business hours is the height of arrogance or ignorance. I’m not sure which one.
i had one app that did not fail at all we have allways noticesd that this app is extrmely fast compered to others
This is usually my case.. theres an app that I have since 2019, and 99% of the times bubble fails, that app runs solid, not this time lol..
I’ve seen this question come up a ton above so just wanted to clarify
It’s a custom message set by our platform engineering team. It’s not standard like a 404 page, for example. It means traffic wasn’t being routed correctly through balancer.
You can read more about the Balancer work from yesterday and today in the link I left at the top of the topic, the “Scheduled Maintenance of Wed, Sept 24” topic.
Hope that helps clarify the message
Same thing. Most of the time, there’s downtime reported here or on status page. Asking my users… and everyone say that it’s working fine. (But not today :P) Even last week, we didn’t got issue
As of today, you have chosen the best and still leading platform. Especially comparing platforms that have been around for years. In its almost 15 years, Bubble has had dozens of supposed superior competitors that didn’t land.
That being said, in my POV, with these actions, Bubble won’t lead for long. Especially without delivering the code and blocking users from hosting on their own like almost all platforms do.
Bubble is not for basic mvp’s and that is not their vision, don’t speculate or spread misinformation
Bubble had to get on to the AI train because if they didn’t, they would certainly be left behind in today’s tech world, unfortunately.
I’m not sure if a lot of people here are unfamiliar with traditional development, so this may blow your mind, but you can work on 2 different things at once. In fact, entire teams can be dedicated to a separate projects.
So while they are working on AI tools, they are also working on upgrading the infrastructure. However things don’t always go smoothly, and this is a result of that. If I remember correctly, upgrading the load balancer should resolve some of the things causing downtime in the past.
In summary: Damned if you do and damned if you don’t.
I agree with you. This unfortunately happens everywhere in the world and in all companies. People always have to look only at the negative aspects.. These are things that can happen, and it must be said that the problem was solved in less than 20 minutes. These things happen to everyone and also to those who develop with code (just think that even AWS can have downtime). It’s technology.
What matters to me is how the problem is handled and how long it will take to solve it. And this time there is little to complain about.. I see a lot saying I’m thinking of leaving.. they’re always the same.. But it is still not clear why they are still here and above all to consult the forum.. Maybe there’s a reason.
However, no one pointed out that yesterday during the maintenance hours absolutely nothing happened…
Let’s try to look on the positive side, sometimes it can really help.
As a matter of fact, our platform engineering team is several X bigger than our AI team. They are working every day to pay down a ton of Bubble tech debt in our infra.
Yesterday’s experiment pushed our new Balancer logic to around 90% of all traffic and worked flawlessly. The team is still investigating what happened today as we ramped up to the same number but it looks like it was caused by an unexpected condition led to all traffic being routed to a test route.
With all that being said, It’s never acceptable to have an outage. And we apologize for the problems today.
The team has been running 90% of load to the new Balancer logic for the past hour and everything is looking healthy. Fantastic news for everyone
Yes but we can’t have engineering focusing on AI while a simple thing don’t work like “Table element” or some random library gets hacked and we need to wait hours to know wtf is happening. I know it’s not your fault, but the tech lead in Bubble.io, CEO’s and many others needs to stop betting in AI and start making simples things works.
Haha, not recursive workflow but Cloudflare was kind enough to inform me that I was DDoS my own app when trying to import 1000 rows.
