The platform has been unstable for a week. Many people reporting similar issues, unfortunately. So it may just be the platform.
Discussion thread:
You can check your Rule/routines and make sure none of them have been damaged. Check the IDE to make sure that everything that is supposed to be scheduled is still scheduled. Remove and re-– pair any problematic devices, then heal the network. It’s a lot of work, but it couldn’t hurt and it might help. On the other hand you might do all of that find it it works OK for a few hours and then breaks again. So some people decided to just not do any maintenance or troubleshooting until the platform is officially stable again. When you have database corruption, anything can happen.