At approximately 4:50PM Easter time, Sago Networks had a utility power drop, which caused our backup systems to kick in. In the process of going through all systems, engineers were alerted to down systems in one row in our dedicated server data center. A main breaker was found tripped in one of the panels in one of our power management modules. Engineers reset the breaker and restored power to the affected racks in the affected row.

 If you are experiencing troubles reaching your server, please contact This e-mail address is being protected from spambots, you need JavaScript enabled to view it or open a ticket through your portal account.

At about 4:30PM EST XO Communication's network start getting major packetloss, particularly in the Orlando Area. We have closed down our XO uplink to prevent customers from traveling across this congested section of the internet. If your services were slow in the past 45 minutes, but now they are back to normal, this was probably the issue. After the packet loss has subsided we will turn the uplink back on and update this page.

At 2:43pm EST our local power company experienced a brief interruption of service lasting two seconds. All automated protection measures enacted, UPS/Battery power kicked in, and a few seconds later the generator automatically started. Facility power was switched to generator. There were no interruptions in power service to equipment anywhere in the facility. The only unexpected event was the breakage of an air conditioning belt on a backup unit. The belt was immediately replaced.

UPDATE 12:30 PM: This morning, we noticed a denial of service attack, that had decent volume, going towards one of our customers. The impact initially of this attack was not that severe, and we were trying to work with the customer to get the attack under control without impacting our customer via our IDS system. After a short time, the attack grew in intensity, and did start to impact other customers on the same network segment. At approximately 9:30AM Eastern, the attack grew again, however started to include our core routers in the attack, which was not seen previously. The attack directed at our core routers caused BGP to begin to flap with our peers to the Internet, as well as with other core routers. Due to the volume of traffic directed to our routers, especially here in Tampa, we had slow responses when consoled in working on the issue. We were able to get some additional blocks and ACLs in place to stop the attack against our core routers. This took time obviously working with slow console response. At approximately 11:00AM Eastern, we were able to restore all BGP sessions and traffic did return to normal.