If it has been so well thought out, then why isn't there a parameter available for each service that says "don't automatically restart this if it fails"?
There are times that services WILL fail, and should not be restarted until the underlying problem is fixed.
And that binary log file? absolutely ridiculous, why? so i have to learn a new tool to find out why a service stopped? When before all i had to do was check the last 20 or so lines of a text file.