Incident postmortem from 11/29

I’m just going to keep repeating this again and again so that Bubble & Co. don’t have to. And understandably, especially in circumstances like this one, it could be received poorly if it came from them. So here it is:

If you have an app that is making money and these kind of issues are hugely detrimental to you or your customers… Please, please please, I beg you, move to a dedicated instance. Either your app makes money or it doesn’t. If it’s making you money then you need to move to a dedicated instance. If it’s not making you money, then these issues can’t be “very serious” they can only be “man that really stinks”.

I understand from a non-technical background this might not make any sense. However, I implore you to go try and spin up a server with all of the necessary dependencies and have someone write your app code. Once they’ve written it, try making updates without bugs and try keeping your server from crashing. It’s not the easiest thing to do and it’s definitely not cheap.

  • Bubble makes no guarantees that your app won’t experience bugs from time to time (go read the ToS)
  • Bubble is working on a solution to this problem for the non-dedicated cluster, but it’s not going to be available anytime soon.
  • Bubble currently has a solution - move to a dedicated instance

For further reading: