We had a script that malfunctioned this afternoon that caused the problem - we got everything back up and running properly at about 5:00 PM.
Now, on to the comments...
100% agreed. There should be a grace period of minimum 15 minutes, ideally at least 1 hour.
I disagree. The more we can "incentivize" providers to keep their feeds running, and to restore their service as soon as possible if they go down, will result in a better user experience.
A good example of why feed providers shouldn't instantly lose their access to the database when a feed goes down.
I'm not sure what you mean by losing access to the database? I'm assuming premium subscription features... in any case, I suspect a 45 min outage didn't cause the world to come to an end because you lost your premium subscription for 45 min. Come on, let's be realistic.
In any case, we're back up, and we're sorry about the outage. With a platform as complex as this, every once in a while we're bound to have a hickup or two.