We've got a system in place to protect against outage, attack and DDOS, but seems there was an error in the process tonight that made part of the issue tonight, and we are checking this to resolve it.
We have identified the problem, and working now on short term, and long term solution.