Welcome to AviationOutlook: The aviation newsletter built exclusively for decision-makers.
Read by the people who run this industry. Become an “annual” member now and get 20% off.
As a paid member, you get access to:
200+ Reports - All major airlines + aerospace & defense companies (updated regularly).
Deep-dive analysis & insights.
Industry outlook reports.
Recommended:
Check Out All Reports
Key Takeaways
A new router at a remote radio site in El Dorado, Arkansas, plus one at a radar site in Amarillo, Texas, flooded the FAA network with 1,000 connection attempts every 4 to 5 seconds.
Instability reached 17 of 22 en route centers, and the ground stop lasted 4 hours 21 minutes.
Root cause is still open: faulty router, installation error, or fragile legacy architecture. Officials report no indication of a cyberattack.
Copper-to-fiber cutovers continue through 2028, so every cutover is a live risk event until then.
The recovery tail cost airlines nearly as many cancellations as the event day itself, and Thanksgiving is nine days away.
Are you a part of a team?
Equip your whole team with aviation industry intelligence. Consider “Group” plan.
What Happened on November 13
On Friday, November 13, 2026, the FAA halted departures across the country after its telecommunications network turned unstable. It was the first nationwide ground stop since 2023.
Controllers and technicians spent about 45 minutes trying to find the cause before the order went out. Departures resumed at 8:15 p.m. Eastern.
Friday, Nov 13, 2026 (Eastern time)
About 45 min Troubleshooting without an identified cause
3:54 pm Nationwide ground stop advisory issued
8:15 pm Departures resume
Ground stop 261 minutes (4 h 21 min)
The 2023 stop, caused by a NOTAM system failure, ran from about 7:20 a.m. to roughly 9 a.m. Friday’s stop lasted more than 2.6 times as long.
Let’s analyze everything in detail.
What the cockpit and the tower experienced
Approximately 3,500 aircraft were airborne when the stop began, and every one landed safely. The FAA says this was not a full outage, and some frequencies and radar feeds stayed reliable.
The controllers’ union president said communications with some planes were disrupted. Controllers worked with one another to find working frequencies and move flights onto them.
Aircraft airborne at order time about 3,500, all landed safely
Facilities with instability 17 of 22 en route centers, plus several TRACONs
Cyberattack indication none reported
How a Router Swap Became a National Event
L3Harris was moving a remote El Dorado radio site from copper lines to fiber and deploying a new router there. When that router came online, a router installed about three weeks earlier in Amarillo also began flooding the network.
The FAA says its legacy equipment became unstable under the data load.
L3Harris blocked the two locations, the network settled, and the broadcast function was disabled on the new routers.
The flood rate in numbers
The FAA’s reported rate converts into a sustained load that older time division multiplexing gear was never designed to absorb.
The NYT account describes the new routers pinging the network hundreds of times per second, which matches the arithmetic below.
Reported flood rate: 1,000 connection attempts per 4 to 5 seconds
Per second 200 to 250
Per minute 12,000 to 15,000
Per hour 720,000 to 900,000
Across the 45 minutes before the stop, if sustained: 540,000 to 675,000
The pattern resembles what network engineers call a broadcast storm, where one device’s traffic multiplies across every connected link. The FAA’s own wording, a broadcasting function that sent data more widely than necessary, points the same direction.
Why one remote site reached 17 centers
The FTI network connects more than 4,400 facilities through shared infrastructure. A fault at the edge therefore travels inward to the hubs.
The FAA administrator told reporters the network will keep breaking until it is replaced. Maintenance logged on paper also slowed diagnosis.
Shared-fate exposure
Facilities on the FTI network more than 4,400
En route centers affected 17 of 22 (77%)
Remote sites that triggered it 2 (El Dorado, AR and Amarillo, TX)
Who Owns the Failure
Nobody has yet established whether the router, the installation, or the architecture failed. The router’s maker told reporters its devices appeared to work as intended.
The Transportation Secretary said officials must determine whether this was a technical install error or an issue in the install itself. Both routers are being examined at an L3Harris lab in Florida, and the FBI is involved.
FAA Legacy TDM equipment became unstable; no sign of cyberattack
Secretary Install error versus router fault still to be determined
L3Harris Supporting the investigation; more than 70% of conversion complete
Router maker Devices appeared to work as intended
Controllers Some aircraft communications disrupted; flights moved to working frequencies



