The November 13 ground stop sits between two risk calls. The one that stopped every departure in the country was right. The other treated a router swap at a remote radio site as a small job, on a network where it was not.
Bryan Bedford, the FAA administrator, described the first call to reporters on Capitol Hill on November 17. He asked the agency’s career staff whether they were “100% certain that we are safe now, as we were two hours ago,” and “I didn’t get a yes.” So, as Fox News recorded it, the FAA decided to “suspend putting more airplanes in the air.”
I hold instrument ratings in airplanes and helicopters, and that is the standard I was trained to: the go needs a yes, and the absence of one is the answer. The system had not gone blind: the FAA says some frequencies and radar feeds stayed reliable. But instability had hit 17 of its 22 en route centers and spread into terminal radar facilities. About 3,500 airplanes already up had to land. Nothing else needed to take off.
Seventeen of twenty-two. At about 3:30 p.m. Eastern, by the FAA’s own account, its “legacy equipment became unstable when data began flooding the FAA FTI network, at a rate of 1,000 connection attempts every 4 to 5 seconds.” L3Harris owns and runs that network under a 2002 contract. The flood traced to two sites: a radio site in El Dorado, Arkansas, where L3Harris had been deploying a new router, and a radar site in Amarillo, Texas, where it had replaced the router three weeks earlier, Reuters reported. L3Harris blocked both, the network settled, and traffic resumed at 8:15 p.m.
Why the router did that is still open. “We’ve been able to recreate what happened in El Dorado,” Transportation Secretary Sean Duffy said on November 17. “But the root cause of what happened, we don’t have clarity yet.” Whatever investigators find inside the box, the number that needs explaining is the reach: two sites, seventeen centers.
It has happened on this network before. On November 19, 2009, 129 of the nation’s air traffic facilities lost FAA telecommunications while Harris was moving services from a legacy network to fiber. The DOT inspector general’s letter reads like a draft of last week. A Harris engineer put a new router in service at Los Angeles carrying a configuration error copied from the router it replaced. The failure spread as a “domino effect,” and about 75 percent of routes stayed blocked for five hours because the tool meant to locate the fault “did not work as intended.”
The letter recorded a fix, too: the new fiber network was “partitioned by en route airspace” to contain “any network anomalies to specific delegated airspace.” That covered the fiber side, then under a tenth of the network’s services. The rest ran on older TDM technology. The FAA says it was legacy equipment that went unstable on November 13, so the investigation should say whether the old side ever got that protection, or whether the flood crossed from new to old.
Remote was how the risk got sized. The FAA’s Q&A asks itself why the swap happened mid-day and answers that maintenance at major facilities “is typically conducted at night,” while this “was a result of a routine replacement at a remote facility.” The AP reported that officials “believed there was a low chance of problems because the work was being done in a remote location.” The best case for that call is the record: by Duffy’s count El Dorado was the 69th of the new routers to go in.
But a clean install shows the swap works. It does not show what a bad one can reach. Amarillo, an earlier swap finished three weeks before, is the other site the FAA named; in 2009 the error sat “dormant” for a month after an install that looked fine. A night window lowers the cost of a failure. It does nothing to its reach.
What GAO put on paper in September. Its audit of the Brand New Air Traffic Control System lists the telecommunications program in one line: “Replace 5,170 copper lines with fiber optic cable,” now due by December 2028, 21 months early. Since the outage the FAA has drawn a line. “It is important to distinguish routine equipment upgrades from the FAA’s broader modernization effort.” Yet Duffy, who told reporters on the night of the outage “I want $50 billion,” described the El Dorado job the next day as changing copper wire to fiber optics, the job GAO’s table names as the rebuild’s telecom program. The two accounts deserve to be reconciled on the record.
The strongest objection is that delay built this hazard, and most of it is right. The routers being pulled, Reuters reported, were Cisco units built in 2001 that the FAA said should have been replaced in 2006. GAO’s own warning points the same way: the net effect of a slow rebuild “could be that air traffic controllers are forced to continue to rely on severely outdated systems that are prone to failure leading to more system outages, ground stops, flight delays,” or worse, safety incidents.
Concede all of it. The transition is the right risk, and taking it means touching a live network thousands of times, where nobody gets certainty. Even in the cockpit, a yes means the run-up checked out: a test you run before you commit. For this network the run-up was written down in 2010. The FAA administrator’s review panel recommended “a functional model of the FAA’s FTI network to simulate and test configuration changes and upgrades,” and the FAA said it accepted the panel’s recommendations. Whether that test bed exists, and whether the El Dorado router went through it, is the first question I would ask.
The ground stop was the margin that held, supplied late and at national scale. Remote is a fact about a building. On November 13, El Dorado reached seventeen centers.
Sources:
- Federal Aviation Administration, Nationwide Ground Stop: Questions and Answers (November 15, 2026).
- U.S. Department of Transportation, Office of Inspector General, letter to the House Transportation and Infrastructure Committee on the November 19, 2009 FTI outage, CC-2010-012 (June 17, 2010), including the enclosed recommendations of the FAA Administrator’s independent review panel.
- U.S. Government Accountability Office, Air Traffic Control Systems: Ambitious New Modernization Effort Needs to Improve Cost and Schedule Planning, GAO-26-107992 (September 14, 2026).
- Fox News, Tech outages live updates (November 17, 2026): Bedford and Duffy on Capitol Hill.
- Reuters, US probes why a router installation sparked mass flight grounding (November 16, 2026, via The Korea Times).
- Associated Press, Scrutiny of FAA intensifies after communications meltdown leads to nationwide grounding (November 16, 2026, via WDBO).
- NPR, FAA blames Friday’s ground stop on ‘equipment issue’ during telecoms network upgrade (November 14, 2026).
- NBC News, Live updates: flights grounded after FAA communications outage (November 13, 2026).