[quote]We are observing particularly frequent and large attacks today at most locations, including Chicago. When an attack is occurring that saturates one of our upstream links, our system instantly null-routes the target to minimize damage, but a small burst of packet loss is seen by a segment of clients (those whose inbound traffic came over a saturated upstream link).
The primary source of these attacks continues to be Amazon, as they still do not have systems in place to effectively mitigate outbound attack traffic. Some other cloud providers that do not proactively notice abuse and do not quickly handle abuse notifications are also involved, including Oracle, Microsoft, and DO.
We will continue to mitigate these attacks as they are seen. We have been in contact with a couple of these other cloud providers in the past about improvements that need to be implemented on their end to help, and we have been assured by some that improvements are on the way. In locations that can receive upgrades, such as Chicago, we are also in the process of those upgrades, to help on our end with attacks that may occur later on.
Attacks activity typically comes in bursts, and holidays are a favorite time for attackers to launch attacks.[/quote]
Apologies for the inconvenience, and as always, thank you for playing on Ephinea!