FAA Tests Two Routers After Network Failure Halted Nationwide Flights for Four Hours
Two routers are now at the center of the investigation into a communications disruption that halted departures nationwide for more than four hours and contributed to delays or cancellations affecting roughly 15,000 flights over two days. The Federal Aviation Administration has not determined whether defective equipment, installation work or another fault produced the network instability.
L3Harris transported the routers from FAA facilities in Texas and Arkansas to its Florida headquarters for deep diagnostic testing, with engineers from the FAA, Juniper and Cisco assisting. Transportation Secretary Sean Duffy said the leading possibilities were “the installation or the router,” but investigators remain open to other findings. Federal law-enforcement partners, including the FBI, remain involved, although officials have found no indication of a foreign cyberattack.
Investigators have narrowed the physical search
The two locations are a radar site in Amarillo, Texas, and a radio communications site in El Dorado, Arkansas. According to the FAA’s official account of the nationwide ground stop, data began flooding its telecommunications network at a rate of 1,000 connection attempts every four to five seconds. The traffic destabilized aircraft communications and radar feeds across much of the system.
The instability reached 17 of the FAA’s 22 Air Route Traffic Control Centers and spread into several terminal radar-approach facilities. This was not a complete loss of radio or radar coverage: some frequencies and feeds remained reliable, allowing controllers to communicate with and track aircraft already airborne. Approximately 3,500 aircraft landed safely while the FAA prevented additional departures.
Isolating the two suspect locations stabilized the network. The broadcasting function on the new routers was then disabled as a precaution, including on approximately 60 similar routers elsewhere in the system. Those actions restored service, but they do not establish whether the originating problem was internal to a router, introduced during installation or tied to how the equipment interacted with the legacy network.
That distinction is what laboratory diagnostics should help resolve. Testing can examine the removed equipment under controlled conditions and compare its behavior with installation records and network data. Investigators must determine why the excessive connection attempts occurred and why instability propagated far beyond two remote facilities. The FAA has also said it will reevaluate when this type of replacement and maintenance work is performed.
A local fault met a national safety threshold
The FAA issued the ground stop at approximately 3:30 p.m. Eastern time on Friday, November 13, and resumed air traffic at 8:15 p.m. The nationwide response reflected the geographic spread and safety significance of inconsistent communications and radar information, rather than a total shutdown of every air-traffic function.
A ground stop reduces the number of aircraft entering a system while controllers safely handle those already in flight. In this case, established procedures preserved critical operations, including certain emergency, law-enforcement and military flights. Duffy summarized the operational standard directly: When we have inconsistencies in our network we’re going to shut it down.
The resulting airline disruption continued well after the network stabilized because aircraft, crews and passengers were displaced across the country. Roughly 15,000 flights were affected Friday and Saturday. By noon Sunday, FlightAware had recorded another 3,798 delays and 1,638 cancellations involving U.S. airports, including 376 delays in Atlanta and 363 delays plus 195 cancellations at Dallas Fort Worth.
The investigation reaches beyond two boxes
The incident occurred during routine work to replace obsolete copper communications links with fiber-optic technology at a remote FAA facility. FAA Administrator Bryan Bedford said that work belongs to the continuing maintenance of the agency’s existing telecommunications network, not the separate, broader air-traffic-control modernization program.
The boundary is important for diagnosing the failure. Replacing an old transmission medium does not automatically replace the wider architecture, its interfaces or every legacy component connected to it. Duffy described the FAA as still operating a telecommunications architecture dating from the 1980s. The agency says legacy equipment became unstable as the excessive connection attempts flooded the network.
The removed routers may identify the initiating fault, but the national consequence also raises a systems-integration issue: how a disturbance associated with one or two remote sites spread into 17 regional control centers. The next decisive result will be whether engineers can reproduce that behavior and separate faulty hardware from installation or configuration work. Until then, the routers are the focus of testing not an established cause.
| More aerospace and engineering stories, right in your MSN feed. Follow AMI on MSN |
By Thomas Caldwell — AMI’s senior editor for mechanical and mobility engineering, covering vehicle electronics, systems integration, electrification, chassis systems, propulsion, and safety policy.
