Skip to the yard
Waxsealai

Legacy Cisco Servers Overwhelmed During U.S. Flight Communications Outage

Shipped
Shipper
[email protected] (Jowi Morales)
Readers
619

A network switchover exposed a weakness in infrastructure dating to 2001, disrupting flight communications for about four hours.

Legacy Cisco Servers Overwhelmed During U.S. Flight Communications Outage

The failure that disrupted U.S. flight communications for about four hours came from infrastructure dating to 2001, according to Transportation Secretary Sean Duffy. His clarification shifts attention from newly installed Juniper routers: he said those devices stayed operational, while older Cisco servers were flooded during a network switchover in El Dorado, Arkansas. The episode shows how newer equipment can coexist with older systems that still carry essential network traffic.

Duffy described the incident in an interview on CNBC. He said the older servers had been out of commission since 2006, but remained part of the network infrastructure. During the switchover, one server or router began flooding the network and caused inconsistent communications. The account does not specify which particular device initiated the flood, or explain why equipment described as out of commission was still part of the system.

How the outage spread

The first reported communication trouble appeared at Indianapolis Center, which monitors airspace across northern Kentucky, southern Indiana, southern Ohio and parts of West Virginia. Staff there had difficulty communicating with flight crews. The Federal Aviation Administration then searched its network for the source and identified an FAA radio site in El Dorado and a radar site in Amarillo, Texas, as possible origins of the problem.

At first, recently installed Juniper routers at those locations appeared to be responsible. FAA technology chief Rebecca Guy said the sites were generating traffic at a high rate: “a thousand connections over every four to five seconds.” That volume might be manageable on many modern networks, but the FAA’s established communications system relies on older copper infrastructure. In this case, the traffic overwhelmed parts of the network and interrupted communications between air traffic controllers and pilots.

The early account focused on a feature in the Juniper routers that was later disabled across the network. Duffy’s subsequent explanation added a different layer: he said the routers had not failed, and that older Cisco servers were flooded during the switchover. The two accounts leave technical details unresolved. In particular, they do not fully explain how the router feature, the switchover and the older servers interacted to produce the network inconsistency.

Why communications mattered

The disruption did not affect the Traffic Collision Avoidance System, which relies primarily on equipment aboard aircraft to communicate with other aircraft transponders. But that protection does not replace coordination between controllers and flight crews. With communications unreliable, the FAA could not coordinate the aircraft in the air with sufficient confidence, so it ordered a nationwide ground stop.

The FAA restored its communications network to normal after about four hours. That marks the duration of the network disruption, not necessarily the time required for travelers to complete their journeys. Cancelled and diverted flights leave passengers waiting while airlines also handle later arrivals and scheduled departures. The resulting delays can persist after the underlying network has recovered.

The Wall Street Journal reported that the incident began after a newly installed router started broadcasting widely across the network. The initial explanation raised questions about installation, configuration and equipment. FAA Administrator Bryan Bedford said the agency was investigating whether the cause involved faulty installation, faulty configuration or a problematic router. Duffy’s later account points to older servers as the equipment that became flooded, though it does not provide a complete technical reconstruction.

Modernization and recovery

The FAA is upgrading its decades-old copper network to fiber, a project described as costing billions of dollars. The incident underscores the operational difficulty of making that transition: new equipment must work with legacy infrastructure while carrying live communications. A device can perform as designed and still contribute to trouble if a feature, traffic pattern or switchover exposes a weakness elsewhere in the network.

The practical questions now concern the system’s dependencies and safeguards. The available account does not say how many older servers remain in use, what protections limit excess traffic, or how future switchovers will be checked for similar effects. Those details matter because replacing network links alone may not address the behavior of older servers or the connections between old and new equipment.

Past disruptions offer a reminder that technical recovery and operational recovery are different tasks. After the 2024 CrowdStrike software outage, airlines including United, Delta and American cancelled flights, and recovery took several days; Southwest was less affected, in part because it used an older operating system. That comparison does not establish how long this incident’s flight delays lasted. It does show why a four-hour communications outage can create a much longer backlog for passengers and airline operations.

The central takeaway is that the newer routers were not, in Duffy’s account, the failed devices. The outage exposed a weak point in older infrastructure during a network change. As the fiber upgrade proceeds, the key measure will be whether the FAA can identify and manage the older systems that remain in the path of critical communications.

Source: [email protected] (Jowi Morales) · www.tomshardware.com

Same vesselRelated cargo