We've noticed that the delete flag in our default downtime model can cause confusion, particularly to new users, whose services get delete unknowingly after 5 days.

Instead of including the delete flag, we will opt to continue polling the service once an hour once a service has been offline for >= 5 days.