The webapp is gone and bors doesn't respond to commands.
Oh, and unlike yesterday's issue this happens (the webapp error) before requiring the user to log in.
Works now.
I appreciate your work on this, but I'd really like to know what went wrong and what was fixed, if you don't mind sharing.
These downtimes are pretty bad for our dev team and I'd like to know if we should consider running our own instance of bors (e.g. would these downtimes be avoidable then?).
I'm pretty sure it's GitHub.com's fault. Their status page reports that they were having trouble with their API ingress yesterday.