Further investigation has revealed a routing-engine process crash on a master unit of an aggregation stack, servicing Internet connectivity to several cabinets within the Chicago Data Center. As the crash happened, the backup routing-engine was automatically brought into service and restored routing functions within a few minutes. We're working with the switch manufacturer to determine the source of the bug and preventative measures through a software patch, which will be applied in a future non-impacting maintenance window. At this time, services to previously impacted cabinets has been stable and there's no concern for any further events.
Posted Dec 14, 2017 - 13:36 CST
Investigating
We're currently analyzing a network event in one aggregation zone of the CHI-3 data center. For servers in this zone, network connectivity on the public network was lost for approximately 2 minutes beginning at 12:53 PM central time. At this time connectivity is restored and functional, but we are still investigating the event to determine the root cause. We will provide further updates as they become available.
Posted Dec 14, 2017 - 13:05 CST
This incident affected: Bare Metal and Dedicated Cloud Services (Chicago).