Live coverage of the FAA nationwide ground stop, 13 November 2026Last checked: Monday 16 November 2026, about 05:30 ET
Home › What Caused the FAA Ground Stop?
FAA Ground Stop · 13 Nov 2026

What Caused the FAA Ground Stop?

The preliminary official account points to a single chain of events: a contractor replacing copper connections with fibre optic cable switched on a new router at El Dorado, Arkansas; the router flooded the FAA's communications network; a second router at Amarillo, Texas, did the same; and 17 of the FAA's 22 control centres became unstable. The root cause — faulty equipment, installation error or configuration — has not been decided. This page gives the confirmed account, flags the two numbers that do not match, and reports the router dispute without taking a side.

In short

According to the FAA's preliminary account, contractor L3Harris was replacing copper with fibre optic cable at a radio site in El Dorado, Arkansas, and deployed a new router. When it was switched on, the router flooded the FAA Telecommunications Infrastructure network with about 1,000 connection attempts every 4 to 5 seconds. A router installed about three weeks earlier at a radar site in Amarillo, Texas, also began flooding the system. Legacy equipment became unstable, 17 of 22 Air Route Traffic Control Centers were affected, and the FAA ordered the nationwide ground stop at 3:54pm ET. The root cause is still under investigation, the router's make and model are disputed, and the count of routers disabled as a precaution differs between the FAA's written account (about 60) and Secretary Duffy (69).

The FAA's preliminary account

The fullest official version of events was published by the FAA on Sunday 15 November, in a questions-and-answers document. It says L3Harris, the contractor modernising the FAA's communications links, had been doing work at the El Dorado site, including deployment of a new router. When data began flooding the FAA Telecommunications Infrastructure network — FTI — at a rate of 1,000 connection attempts every 4 to 5 seconds, legacy equipment became unstable. The FAA identified two potential locations: a radar site in Amarillo, Texas, and a radio communications site in El Dorado, Arkansas. Secretary Duffy added detail in briefings: the work at El Dorado was part of replacing copper wire with fibre optic cable, and when the new router was switched on, a router installed about three weeks earlier in Amarillo started to flood the system as well. The instability reached 17 of the FAA's 22 Air Route Traffic Control Centers and spread into several Terminal Radar Approach Control facilities. The FAA describes this as its initial investigation. It is not a final root-cause report, and no date has been given for one.

Copper to fibre: the upgrade in question

The work at El Dorado was part of a long-running programme to replace the FAA's copper telephone-era connections with fibre optic cable. L3Harris says it has completed more than 70 percent of the copper-to-fibre conversion. The programme as a whole is large: about 2,560 of 5,170 copper connections had been replaced, according to reports of the modernisation effort. The El Dorado job was routine modernisation work of that kind — a site upgrade, not an emergency repair. That is one reason the timing became a question: the FAA's own questions-and-answers document includes the question of why the router replacement work was done in the middle of the day rather than at night. The FAA's answer, and the working practices behind it, are part of what the investigation is examining. What is not in dispute is the sequence: the new router was switched on during the day, the flooding began, and within about half an hour the nationwide stop was ordered.

How one router grounded a country

The mechanism is a network flood. A router that repeatedly attempts to connect — about 1,000 attempts every 4 to 5 seconds — can overwhelm the equipment around it, and the equipment that fails can then pass the problem on. The FAA's network links control centres to each other and to the radio and radar sites they use. When the flooding began, legacy equipment in the network became unstable, and the instability spread from centre to centre until 17 of 22 were affected. The system was designed in an era when a single site failure was expected to stay local. The flood did not stay local, because the network itself carried it. That is why a fault at two sites in Arkansas and Texas could ground departures in New York, Chicago, Los Angeles and Miami within the hour. The FAA's response was to disable the broadcast function on the routers involved and on similar routers across the system — a precaution that also reveals how widely the same equipment is deployed.

The router count that does not match

After the stop, the FAA disabled the broadcast function on similar routers as a precaution. How many is not agreed. The FAA's written questions-and-answers document says approximately 60 similar routers. Secretary Duffy, appearing on ABC's This Week on Sunday 15 November, said the broadcast capability of 69 routers had been disabled, and described it as all 69 routers of that type. The two figures have not been reconciled. This site reports both, because picking one silently would mislead readers. The difference may be a matter of counting — routers of that type against routers with that function enabled, for example — but neither the FAA nor the Department of Transportation has explained it. Until they do, both figures stand.

The router make and model: a live dispute

The make and model of the router is disputed, and the FAA itself names no vendor. The Wall Street Journal reported that the El Dorado router was a new Juniper Networks device installed by L3Harris. HPE, which acquired Juniper in July 2025 in a 14-billion-dollar deal, said its devices did not fail and appeared to have worked as intended. An earlier account sourced to CBS blamed old Cisco 600-series routers dating from 2001 to 2006. That account is unconfirmed and is contradicted by the Journal's reporting. The FAA's questions-and-answers document names no hardware vendor at all. Both suspect routers were removed and are under deep diagnostic testing at an L3Harris laboratory in Florida, with FAA, FBI and other law enforcement personnel and engineers from Juniper and Cisco assisting. Until the testing reports, no vendor should be stated as fact — and this site does not.

Faulty equipment, installation error, or configuration?

Secretary Duffy's own words show the question is open. He said the investigation would determine whether it was faulty equipment or contractor error, and later said he thought it was the installation or the router that was the bad actor. He also asked publicly whether L3Harris had shown incompetence and whether the company didn't meet the moment. L3Harris says it is supporting the investigation. The three possibilities point in different directions: a faulty device would raise questions about procurement and testing; an installation error would raise questions about contractor procedure and FAA oversight; a configuration mistake would raise questions about both. The distinction matters for what gets fixed — and for who is held responsible — which is why the investigation's answer, when it comes, will matter more than the speculation now. No root-cause report date has been given.

How the FAA says it will prevent a recurrence

The FAA's questions-and-answers document addresses prevention directly. The immediate step was disabling the broadcast function on the routers involved and on approximately 60 similar routers across the system. The longer-term answer is the modernisation programme itself: replacing the copper-era equipment with fibre and new systems, funded by 12.5 billion dollars approved in July 2025, with Secretary Duffy now seeking a total of 50 billion dollars. But the ground stop also raises an awkward question for that programme: the failure happened during modernisation work. The FAA's answer is that the upgrade did not cause the failure in the sense of a design flaw — the failure was in the equipment and the installation — and that the upgrade is what will eventually remove the legacy equipment that proved vulnerable. Critics, including House Democrats, ask why only a fraction of the money already provided has been spent, and whether the programme is being managed well enough to prevent the next failure. Duffy himself has warned that without the new system, more failures of this kind will happen.

Frequently asked questions

A communications failure in the FAA network. Router work during a copper-to-fibre upgrade at El Dorado, Arkansas, and Amarillo, Texas, flooded the network with about 1,000 connection attempts every 4 to 5 seconds, destabilising 17 of 22 control centres. The root cause is still under investigation.

L3Harris is the contractor modernising the FAA's communications links, replacing copper connections with fibre optic cable. It says it has completed more than 70 percent of the conversion and is supporting the investigation.

That is disputed. The Wall Street Journal reported a new Juniper device; an earlier CBS-sourced account blamed old Cisco 600-series routers. The FAA names no vendor. HPE, which owns Juniper, says its devices did not fail.

The FAA's written account says approximately 60 similar routers. Secretary Duffy said 69 routers of that type. The discrepancy has not been reconciled.

Undecided. Both suspect routers are under diagnostic testing at an L3Harris laboratory in Florida. Secretary Duffy has said he thinks it was the installation or the router that was the bad actor.

That question is in the FAA's own questions-and-answers document. The working practices behind daytime router replacement are part of what the investigation is examining.

Immediately, by disabling the broadcast function on the routers involved and on similar routers system-wide. Longer term, through the air traffic control modernisation programme, for which Secretary Duffy is seeking 50 billion dollars.

Sources & check time. FAA, Nationwide Ground Stop: Questions and Answers (15 Nov 2026, read directly 16 Nov); Travel Weekly Australia; TechTimes; Reuters via BusinessWorld; Wall Street Journal via Yahoo/Men's Journal; IANS/HiIndia. Last checked about 05:00 ET, Mon 16 Nov 2026.
Figures are given with their snapshot time and source. Developing items are marked as developing. Last checked Monday 16 November 2026, about 05:30 ET.